CAFC Rules “Secret Prior Art” Requires Only Appreciation that Invention Made in Teva Pharmaceutical*

By Eric Guttag
December 4, 2011

The novelty bar under 35 U.S.C. § 102(g)(2) has been characterized (accurately) as involving “secret prior art.” By “secret,” I mean that the patent applicant likely won’t even be aware of that “prior art” because it hasn’t been published (or is otherwise not publicly available to see). In fact, since the 1973 CCPA case of In re Bass, 474 F.2c 1276, 35 U.S.C. § 102(g)(2) (and its predecessor) has also been an annoying source of “prior art” for the purposes of obviousness under 35 U.S.C. § 103(a).

While having “secret prior art” undercut an effort to patent an invention may be frustrating for many patent applicants, yet another source of irritation is the doctrine of “inherency.” For example, the prior efforts of others can be invalidating “prior art” whether or not those “others” recognize the significance or relevance of the properties of that “prior” invention. (This frequently happens in the technical field of chemical compounds, and especially, drugs.) In fact, all that is required is that the “others” creating that “prior” invention appreciate that, in fact, they made it.

The doctrines of “secret prior art” and “inherency” both merged in the case Teva Pharmaceutical Industries Ltd. v. AstraZeneca Pharmaceuticals LP to surprise, and unpleasantly upend the patentee. In Teva Pharmaceutical, the claimed invention covered by Reissued U.S. Pat. No. 39,502 (the “’502 patent”) was stabilized pharmaceutical formulations for treating dyslipidemia (i.e., an abnormal amount of lipids in the blood). The active in these pharmaceutical formulations was a statin drug. But the critical aspect of the claimed invention was the use of an amido-group containing polymeric compound (“AGCP compound”) as the only component to stabilize the statin drug (i.e., no other stabilizers are used other than the claimed AGCP compound). Also important to know (as will shortly become apparent) is that the ‘502 patent claimed the priority benefit of a provisional application filed on April 10, 2000.

In October 2008, Teva Pharmaceutical sued AstraZeneca for infringing the ’502 patent based on the manufacture and sale of AstraZeneca’s well-known CRESTOR® drug for treating high cholesterol. CRESTOR® is a stabilized statin (rosuvastatin calcium) formulation which may also be used to treat dyslipidemia. The designed for CRESTOR® was tribasic calcium phosphate, which is not an AGCP compound. But CRESTOR® additionally contained crospovidone, which was an AGCP compound, and would “inherently” act as a stabilizer for the statin active. AstraZeneca included crospovidone in CRESTOR® expressly as a disintegrant, but did not understand that crospovidone also had the stabilizing effect for the statin active.

AstraZeneca moved for summary judgment of invalidity under 35 U.S.C. § 102(g)(2), alleging that it had conceived and reduced to practice its CRESTOR® drug to practice prior to Teva Pharmaceutical’s first conception of the claimed AGCP compound stabilized statin formulation. One of these earlier conception/reduction to practices was based on the “undisputed showing that, in mid-1999, [AstraZeneca] manufactured a 10,000-unit batch of a rosuvastatin calcium formulation containing the same ingredients in the same amounts as its commercial [CRESTOR®] drug.” (Interestingly and certainly tactically, AstraZeneca also conceded infringement of the asserted claims of the ‘502 patent for purposes of this summary judgment motion, even though the inclusion of tribasic calcium phosphate, a non-AGCP-compound, as the stabilizer in its CRESTOR® drug might have also been a strong noninfringement defense.) The district court granted AstraZeneca’s summary judgment motion, and held that the asserted claims in the ‘502 patent were invalid under 35 U.S.C. § 102(g)(2) in view of AstraZeneca’s prior conception/reduction to practice of its mid-1999 rosuvastatin calcium formulation (and potentially other prior conception/reduction to practices of these 1999 formulations as will be mentioned later).

On appeal, Teva Pharmaceuticals primary argument was that the district court misapplied 35 § 102(g)(2) by “failing to require AstraZeneca to prove that it appreciated the stabilizing effect of crospovidone in its drug formulation.” That argument went nowhere with the Federal Circuit panel (Judge Linn writing the opinion, with Chief Judge Rader and Judge Dyk joining). While it was undisputed that AstraZeneca “did not understand that crospovidone acted as a stabilizer in its drug prior to Teva [Pharmaceutical]’s conception, if at all,” that did not end the applicability of 35 § 102(g)(2).

Instead, citing first the 1972 CCPA decision of Parker v. Frilette, Judge Linn’s opinion ruled that “[a]n inventor need not understand precisely why his invention works in order to achieve an actual reduction to practice.” Even more on point was the 2001 Federal Circuit decision in Dow Chemical Co. v. Astro-Valcour, Inc. (“undisputed, clear and convincing evidence in the record that [the alleged prior inventor’s] employees immediately appreciated what they had made, and indeed its significance, when they made [the claimed] isobutane-blown foam”) which held that the first to invent need not be “the first to appreciate the [patentable aspect] of the invention.”

Judge Linn’s opinion for the Federal Circuit panel also cited the 2001 Federal Circuit decision in Mycogen Plant Science v. Monsanto Co., and the 2005 Federal Circuit decision in Invitrogen Corp. v. Clontech Labs., Inc. which, like Dow Chemical, held the prior inventor need only “prove that it appreciated what it had made,” and not “everything about how or why its invention worked.” Relying upon the collective holdings in Dow Chemical, Mycogen Plant Science, and Invitrogen, the Federal Circuit panel then concluded that “it is apparent that the district court correctly entered summary judgment” of invalidity of the asserted claims in the ‘502 patent under 35 § 102(g)(2).

Others have taken the view that, with the change to “first to file” regime which will come into effect on March 16, 2013 under the “new” § 102 provision of the America Invents Act (AIA), “prior art” assertions like those made by AstraZeneca under 35 § 102(g)(2) in the Teva Pharmaceutical’s case will be “eliminated.” Au contraire, “old” § 102(g)(2) “prior art” may still be alive and well, even in view of the AIA (or as I call it, the Abominable Inane Act). As of March 16, 2013, the “new” § 102 of the AIA bars patenting if the claimed invention was “in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.” AstraZeneca might still be able to assert some other “prior art” evidence, namely, it’s late summer-1999 CRESTOR formulation (which were characterized in the Federal Circuit panel opinion as being “disclosed”) under the ““new” § 102 as either a “public use” or at least “otherwise available to the public.” This other evidence would be still prior to Teva Pharmaceuticals earliest effective filing date (i.e., the provisional filing date of April 10, 2000).

Please also note that the “new” § 103 of the AIA still says “prior art” without saying that “prior art” has to be one of the barring acts described in the “new” § 102. That “prior art” in the “new” § 103 could still be prior inventions by others prior to the effective filing date of the patent/patent application, even if supposedly “secret.” Put differently, those reports of the supposed “demise” of “old” § 102(g)(2) in view of the AIA may be greatly exaggerated.

*© 2011 Eric W. Guttag.

The Author

Eric Guttag

Eric Guttag  
Mr. Guttag has over 38 years of corporate and private intellectual property law experience on patent, trademark, copyright, trade secret and unfair competition matters, computer and Internet law, including patent application drafting, prosecution, and patentability studies; infringement and validity studies; international patent prosecution; patent and know-how licensing; consulting, confidentiality, clinical study and research agreements; trademark searches and opinions; trademark registration and prosecution; trademark freedom-to-use studies and trademark litigation and dispute resolution.

Warning & Disclaimer: The pages, articles and comments on do not constitute legal advice, nor do they create any attorney-client relationship. The articles published express the personal opinion and views of the author as of the time of publication and should not be attributed to the author’s employer, clients or the sponsors of Read more.

Discuss this

There are currently 10 Comments comments.

  1. Paul F. Morgan December 5, 2011 10:42 am

    The business and investment uncertainty of 102(g) secret prior art, and even secret prior patent rights thereunder, is a key reason it is being eliminated under the AIA.

    Also, the AIA does not just eliminate 102(g) it also simultaniously re-defines what IS prior art in a new 102. Thus I do not see how old [present] 102 definitions of what is prior art could still be prior art under 103, especially expressly eliminated 102(g)? What is prior art under 103 is what is defined as prior art in 102.

  2. Paul F. Morgan December 5, 2011 10:51 am

    P.S. Also note that “date of invention” was changed to filing date in AIA 103 itself [for joint research provision], and elsewhere in the AIA.

  3. EG December 5, 2011 10:59 am

    “What is prior art under 103 is what is defined as prior art in 102.”

    Paul, you could be correct if the courts interprete “new 103” under the AIA that way. But remember that even “new 103” under the AIA (like “old 103” under the existing statutes) says “prior art” without defining that “prior art” to mean only barring acts described under “new 102.” At this point, until the courts say otherwise, I would be careful in simply assuming that the the breadth of “prior art” in “new 103” is limited by those barring acts described in “new 102.”

  4. Paul F. Morgan December 5, 2011 3:07 pm

    Eric, this thread leads to an interesting question as to what “prior art” has ever been successfully asserted under 103 that is NOT based on 102? I am not aware of any, off hand. If there are any, that would be valuable for interpresting the AIA. At least one missguided Fed. Cir. panel decision and treatise writter does seem to call a prior secret commercial use a 102 “public use” by [incredibly] ignoring the word “pubic.” in 102. But in fact, as clearly set forth by Judge Learned Hand in Metallzing Engineering [the source of the secret commercial use doctrine by that name subsequently favorably cited by the Fed. Cir. and the Sup. Ct.] it is only a purely personal “forfeture” [thus equitable in nature] and the Fed. Cir. in Gore v. Garlock, etc., makes that very clear – a secret commercial use is NOT a bar or prior art against anyone other than the secret prior commercial user itself.

  5. Paul F. Morgan December 5, 2011 3:19 pm

    P.S. New AIA 103 starts off with: “A patent for a claimed invention may not be obtained though, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, …”
    I.e, a specific reference to new, AIA, 102.

  6. Anon December 5, 2011 3:59 pm


    Has not it been stated elsewhere that Metallazing Engineering has been tossed with the re-write of 102?

    That the fact that personal and non-personal aspects have been combined into into one section means that secret commerical use is no longer only a purely personal “forfeiture” [thus no longer merely equitable in nature] and is, in fact, definitely a bar against everyone?

    The new 102 just does not make that distinction between personal and public bar, but rather treats them the same. I think this is where some dervie that the exceptions to 102 must now be personal exceptions.

    Either way, though, I think that one cannot simply pick and choose a little of either (much like being partially pregnant – you either are, or you are not) – either the bars and exceptions will be personal, or the bars and exceptions will not be personal. The stronger the argument one makes for not having the exceptions personal means the bars cannot be so. Conversly, the stronger one argues that the bars must be personal, so too must be the exceptions.

  7. EG December 5, 2011 4:10 pm

    “New AIA 103 starts off with: “A patent for a claimed invention may not be obtained though, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, …”
    I.e, a specific reference to new, AIA, 102.”

    Point well taken, Paul. You can only hope the courts will read new 103 that way, and ditch the concept of “secret prior art” under “old” 35 USC 102(g)(2). (I also am unware of any case under “old” 103 that treated as “prior art” that which isn’t denoted as a bar under “old” 102). But then again, I and the Federal Circuit (in Waymark) thought Deepsouth Packing was overturned by 35 USC 271(f) as far as “extraterritorial effects” but SCOTUS then said otherwise in Microsoft v. AT&T.

  8. Paul F. Morgan December 5, 2011 4:31 pm

    Anon, that is a real sore subject of debate – whether or not secret commercial use will be eliminated as any obstacle to getting a patent or not. 102(a)(1) is not itself that clear as to whethor the end phrase “or otherwise publicly disclosed” is a catch-all or a condition for everything previously recited. Importantly, since the Metallizing Engineering doctrine is equitable rather than statutory, as noted above, and not specifically expressly overuled, why should it be treated as clearly removed by the new statute?
    Futhermore, there is very strong public policy expressed by the Sup. Ct.. the Fed. Cir. and many others against tacking a long patent term onto a long trade secret protection term for IPL “double dipping.”
    Do you really think the Federal Circuit or the Supreme Court is going to want to uphold a patent from an AIA application filed after March 2013 where the applicants admit the patent is on an invention that the applicants or their assiginee had already had in secret commercial use for 15 years before their filing date?
    [Given this ongoing dispute over secret prior commercial use, it might be IC-dangerous not to disclose it in the application? ]

  9. Anon December 5, 2011 6:46 pm

    why should it be treated as clearly removed by the new statute?

    The cleanest answer I have to that is that the entire statute has been rewriten and the multiple parts have been put together in one area, which signifies one treatment for all.

    Do you have any propsition for not treating them thusly? How do you get someone “partially pregnant?

    As for what the Sup. Ct., the Fed. Cir. and many others against tacking a long patent term onto a long trade secret protection term have actually said, all have said “play by the rules Congress sets”. Deference to the RUles of Congress has been a key underlying theme throughout all of the discussions. If Congress now (in it’s wisdom, or lack thereof) have reset the rules, the courts server their position by merely interpreting what Congress has put forth. In other words, you rely too heavily on the courts, when it is Congress that makes the law.

    As they say, “It is what it is.”

  10. EG December 7, 2011 7:57 am

    “The business and investment uncertainty of 102(g) secret prior art, and even secret prior patent rights thereunder, is a key reason it is being eliminated under the AIA.”

    Paul, I have a favor to ask: can you point me to where I can find the legislative history for the AIA that supports your statement? That would certainly be convincing proof (even though Justice Scalia says thtat legislative history is irrelevant v. “plain meaning”) that 102(g) art will be dead under the AIA. Another inference in the AIA that 102(g) art is dead for 103 purposes is the new expanded prior user rights provision (discussed in a more recent thread by Gene) that would make no sense if 102(g) art were still alive under the AIA. Thanks.