Litno454azn

C.D. CRARY & CO.


Calvert D. Crary
May 10, 2010
(914) 763-6520

ASTRAZENECA’S CRESTOR PATENT LITIGATION

We read the post-trial briefs in AstraZeneca’s and Shionogi’s patent infringement case against Teva
Pharmaceuticals, Mylan Labs, Par Pharmaceutical, Aurobindo Pharma, Sun Pharmaceutical,
Cobalt Pharmaceuticals and Apotex in AstraZeneca’s defense of Crestor, its anticholesterol drug. The
case is pending before Judge Joseph Farnan of the U.S. District Court in Delaware, who has historically
been a difficult judge for generics even when the merits are in their favor. Moreover, the primary issue in
this case is inequitable conduct by Shionogi, the patentee, and since inequitable conduct cases are
strongly disfavored by the Federal Circuit, it has always been clear that generics need a nearly flawless
position in order to prevail before Judge Farnan on this ground.
In our analysis, even though the generics presented a powerful case on the inequitable conduct
issue, we think that Judge Farnan will ultimately rule that their proof on the intent element of
inequitable conduct was insufficient and that therefore AstraZeneca and Shionogi will win on this
issue. However, we think that Judge Farnan will probably rule against AstraZeneca and Shionogi
on a second issue in the case, namely the issue of whether the claims in the reissued patent are valid.
If we are right about this ruling, then AstraZeneca and Shionogi will lose their patent rights
altogether on Crestor and the product will go generic, although there will be no opportunity for
follow-on antitrust cases by plaintiff purchaser groups.
The patent asserted against the generics in this case is U.S. Patent Reissue No. 37,314, which was a
reissue of U.S. Patent No. 5,260,440, which was filed for on June 12, 1992 and was issued on November
9, 1993. None of the claims of the ‘440 patent were limited to rosuvastatin itself, the active ingredient in
Crestor, but instead all of the claims covered groups of compounds, some of which included rosuvastatin.
In the reissued patent, all of the claims of the ‘440 patent were canceled, replaced by new “bullet” claims
that were more specifically directed to rosuvastatin itself.
The relevant facts of this case are extensive, but we tried to keep our description of them to a minimum.
Rosuvastatin was a relative latecomer to statins, and by the time Shionogi had gotten to work on the
subject, the general format of a statin and the characteristics of desirable statins were well known. In fact,
the market was quite competitive, and Shionogi was concerned that if Bayer saw the Shionogi patent
application, Bayer would claim rosuvastatin itself by adding it to its own pending application, discussed
below, which had broad “genus” claims. Shionogi avoided tipping off Bayer by claiming only genus
claims in its ‘440 patent application, even though it knew it had selected rosuvastatin for development.
The inequitable conduct issue arose because Claim 1 of the ‘440 patent overlapped with a prior art
“Sandoz” reference, making it highly material to prosecution of the ‘440 patent application. Shionogi’s
C.D. Crary & Co., Building 4, 14 Chapel Road, Box 192, Waccabuc, NY 10597 (914) 763-6520 Fax: (914) 763-6915 www.litigationnotes.com
C.D. Crary & Co. is the business name for Calvert Crary, an independent provider of investment research. Any recommendation contained in this report may not be suitable for all investors.
Moreover, although the information contained herein has been obtained from sources we believe to be reliable, its accuracy and completeness cannot be guaranteed. In addition, the author may
have positions and effect transactions in the securities or options on the issuers mentioned herein and may serve as directors of such issuers, although if so, then any such position or relationship
will be disclosed.  2010 by Calvert Crary.
patent lawyers in Japan became aware of its high materiality, if they were not aware already, because the patent office in the EU had marked it with an “X” on a “European Search Report,” indicating its high level of materiality even if not combined with any other reference. It is undisputed that the Japanese patent lawyers were aware of both the Sandoz reference and the European Search Report, since there is an extremely clear trail of the movement of both documents inside Shionogi. To make matters worse, there is a damaging sequence of events that raises the specter of intentional concealment of the documents. The initial patent lawyer working on the application, Ms. Kitamura, left the company in July 1992, shortly before the information disclosure statement was due at the USPTO. Her replacement, Mr. Shibata, admitted that he was aware of the Sandoz and Bayer references as well as the European Search Report, and was aware of their materiality, but he made no effort to insure that they were disclosed to the USPTO. Instead, he asked a very young department member, Ms. Shimizu, to prepare a response to a USPTO office action in September 1992, and then in December 1992 he reassigned the application again to a Mr. Tamaki. Then, in early 1993, Shibata divided the responsibility for the application, in conflict with normal Shionogi policy, putting Tamaki on the European application and assigning the U.S. application to Shimizu. None of these individuals ever disclosed the references to the USPTO, but it is easy to suspect that the reassignments were made for the purpose of insulating the application from personnel who perceived the need to disclose the Sandoz and Bayer references. AstraZeneca’s explanations for the failure to disclose these highly material references to the USPTO range from “confusion and chaos” to “too busy” to “I thought it had been submitted already.” Since a specific intent to mislead or defraud the USPTO must be shown by clear and convincing evidence in order to establish inequitable conduct, AstraZeneca argues that fraudulent intent has not been shown since its own explanation is at least equally likely. In our view, based on our past experience observing Judge Farnan, we think he is likely to rule in favor of AstraZeneca on the inequitable conduct claim in this case, although it is possible that he will concede that the conduct of the Shionogi personnel is “troubling.” His basis for rejecting the inequitable conduct claim, if we are right, will be that fraudulent intent has not been shown by clear and convincing evidence. On the validity of the patent reissue, however, we are thinking that even Judge Farnan will come out from under his desk and rule in favor of the generics, since the evidence is clear to us. However, knowing Judge Farnan’s history, we recognize that we are sticking our editorial neck way out. The present case is surprisingly similar to the case we discussed in our last issue, cf. Litigation Notes, May 4, 2010, in which Judge Jose Linares of the U.S. District Court in Newark (also a favorable forum for drug companies) ruled that Merck’s reissued claims were invalid because there was no error in the original claims as drafted. In that case, Merck’s patent lawyer who drafted the original claims had said that her “error” was having inadvertently neglected to include “bullet” claims for the specific compound, ezetimibe, in the original patent. Judge Linares ruled that the bullet claims added in the reissue were invalid since there was nothing wrong with the existing “genus” claims that Merck had elected to pursue. In the present case, the same failure to include bullet claims in the original ‘440 patent occurred, but there was nothing wrong with the genus claims that were ultimately included in the patent. AstraZeneca argues that in this case, the genus claims of the ‘440 patent were potentially invalid in light of the Sandoz and Bayer references, but it is quite clear that these claims would have been upheld as valid “selection” claims, since it is well established that a subset of useful compounds within a broad class of prior art compounds is patentable, and there was no error that rendered the prior genus claims invalid. In fact, Shionogi’s U.S. lawyer advised Shionogi in 1998, during the negotiations with AstraZeneca, that Claim 1 of the ‘440 patent was not per se invalid in light of the Sandoz reference because “even if there is some overlap, it can be said that ours is a selection invention.” Further, we are persuaded by the evidence that Ms. Kitamura did not simply forget to include the bullet claims, but rather that the omission was intentional in order to conceal Shionogi’s interest in rosuvastatin from Bayer. AstraZeneca also argued that the evidence fails to establish that Shionogi’s omission of claims specific to rosuvastatin was deliberate. The generics respond that none of AstraZeneca’s witnesses provided any testimony at all as to why the bullet claims for rosuvastatin were omitted, even though the two witnesses who could have rebutted the argument, Kitamura and Shibata, both testified at the trial. AstraZeneca next argues that concealing rosuvastatin from Bayer could not have been Shionogi’s motive since rosuavstatin was included as one of the examples in the ‘440 specification, and Bayer would have seen the examples. The generics counter by saying that describing a compound in an example in a patent specification is nowhere near as big a flag as describing a compound in a claim, especially if it is the only compound so described. Still another argument from AstraZeneca is that one of the grounds for reissue under 35 U.S.C. §251 is satisfied when the basis for seeking narrowing reissue is the belated discovery of partially-invalidating prior art. AstraZeneca then argues that the belated appreciation of known invalidating prior art is the same as the belated discovery of invalidating prior art. The generics respond to this argument by pointing out that here, the evidence is clear that Shionogi’s patent personnel understood the significance of the Sandoz reference well before the ‘440 patent issued. In Merck’s case, the consequences of invalidating the reissued claims was not too severe, since Merck is still alive with genus claims on Zetia in its case against Glenmark Pharmaceuticals, but in the present case, Shionogi surrendered all of the ‘440 patent’s claims in order to obtain the ‘314 patent reissued with bullet claims on rosuvastatin. That, of course, is all that AstraZeneca needed to insure that Crestor has patent coverage, but if these bullet claims are invalid, as we believe, then AstraZeneca’s patent coverage for Crestor will be gone.

Source: http://www.litigationnotes.com/litno454azn.pdf