Takeda committed inequitable conduct to secure a patent on the “me-too” compound lansoprazole.¹¹ In trying to establish the patentability of lansoprazole over the structurally similar, “benchmark” industry standard compound omeprazole, Takeda disclosed a single piece of favorable IMAU test data that purported to show lansoprazole was greater than 20-times more potent than omeprazole. Takeda made the greater than 20-times superior representation because Takeda questioned whether lansoprazole could be patented given that its “[s]tructure is too similar to Omeprazole.” (A1415-19.) Takeda had every reason to expect that it could avoid an obviousness rejection only by asserting that its “me-too” omeprazole analog was significantly more potent than omeprazole itself. Takeda's strategy succeeded. The examiner accepted this assertion and allowed the claim to lansoprazole to issue.

The evidence of record, however, clearly and convincingly showed that Takeda failed to disclose to the PTO a wide range of other test data in its possession that directly contradicted the claim that lansoprazole was greater than 20-times superior to omeprazole. Only one set of data taken in one particular animal testing model, the IMAU model, supported the greater than 20-times superior assertion. Takeda withheld anti-ulcer data that showed lansoprazole to be only slightly more potent than omeprazole. Takeda also withheld data taken from other pertinent test models that showed lansoprazole and omeprazole to be equally potent. And Takeda withheld still more data that showed that omeprazole was the more potent compound. All of these tests directly or indirectly focused on whether a Proton Pump Inhibitor could inhibit the Proton Pump and reduce production of acid in the parietal cell. A reasonable patent examiner would have considered all of the withheld, contradictory data important when determining whether lansoprazole was patentably distinct over the prior art.

The District Court, however, erred by concluding that the withheld data had only a “low level of materiality.” This error resulted from three separate underlying errors, each of which compels reversal:

First, the District Court erroneously held that the withheld data was not inconsistent with Takeda's disclosures to the PTO because Takeda indicated that lansoprazole was 1.5 to more than 20 times more effective than omeprazole. This was clear error. Takeda asserted that the claimed compounds as a whole were between 1.5 and 20 times superior to the prior art; however, the only data that Takeda disclosed comparing lansoprazole to omeprazole -the ID50 values for the two compounds – indicated that lansoprazole was greater than 20-times superior to lansoprazole.

Second, the District Court erroneously concluded that the full range of testing data – including the withheld, entirely contradictory data – would not have been important to an examiner. This was clear error. Every expert to testify at trial agreed that the full range of data was important to an evaluation of the compound, and a reasonable examiner would have wanted to know that most of the data in Takeda' s hands contradicted the claims of greater than 20-times superiority.

Third, the District Court erred as a matter of law by concluding that the level of materiality of the misrepresentation of superiority, and the materiality of the data that was withheld, depended on what stage of patent prosecution the misrepresentation occurred. The District Court lowered the materiality of the withheld data because the greater-than-20-times-superior data was disclosed in the original application, and not disclosed only in response to an examiner rejection or other office action. This is legal error and compels reversal.

The District Court also correctly found that the assertions and selective disclosure of data was “self-serving,” but the erroneous finding of a “low level of materiality” tainted the remainder of the inequitable conduct analysis. Had the District Court properly found a high level of materiality, the record provided compelling support for an inference that Takeda intended to deceive the PTO – an inference sufficiently strong to compel a finding of inequitable conduct. Takeda internally questioned whether lansoprazole was patentable due to its structural similarity to omeprazole. (A1415-19.) The Takeda scientist who was involved in the selection of data admitted that he chose to include only data from the IMAU model in the patent application because of the results – because lansoprazole “showed strong activities in this model.” (A899 (1569:9-19).) Further, Takeda internally considered the withheld, contradictory data to be sufficiently material that it submitted this withheld data (and only the withheld data) to the FDA in its IND and reported it in peer-reviewed journal articles. And Takeda failed to bring a single live witness to the trial to explain why Takeda chose to withhold data from the PTO.

The District Court did not abuse its discretion in finding no clear and convincing evidence of inequitable conduct. Significantly, the District Court found that intent to deceive the PTO was lacking because:

• “Takeda has presented evidence that [it] placed a heavy reliance on the IMAU testing, as compared to other testing methods.” A45, ¶ 75.

• “Takeda itself viewed the IMAU results as the best evidence of the activity of lansoprazole.” A47.

• “Takeda identified the relevant prior art … to the PTO, and did not manipulate the data it reported to the PTO,” and there were “indicia of Takeda's good faith.” Id.

• Takeda “made no misrepresentations to the PTO.” Id.

• Thus, Takeda and TAP affirmatively demonstrated that Takeda's employees involved in the ‘098 patent's prosecution did not intend to deceive the PTO. A45-47, ¶ 74-77.

These factual findings are firmly supported by the record and are not clearly erroneous. On appeal, Teva does not challenge any of these factual findings or assert that they are clearly erroneous.

The District Court correctly concluded that this case was similar to Takeda Chem. Indus,, Ltd. v. Mylan Labs., Inc., 417 F. Supp. 2d 341, 390 (S.D.N.Y. 2006), affd sub nom. Takeda Chem. Indus., Ltd. v. Alphapharm Pty., Ltd., 492 F.3d 1350, 1354 (Fed. Cir. 2007). A47. The Takeda v. Mylan court, and subsequently this Court on appeal, found no intent to deceive the PTO where, as here, “Takeda … relied upon the data in [its internal] report in making critical business decisions,” thus finding “it was entirely appropriate for Takeda to rely on that same data in prepar
ing Table 1 for the PTO.” 417 F. Supp. 2d at 390. No meaningful differences exist between the facts here and the facts in Takeda v. Mylan, and Teva makes no argument distinguishing Takeda v. Mylan in its brief.

On appeal, Teva contends that the District Court committed clear error on three findings of fact underpinning its low-materiality finding. Appellants' Br. at 28-29. Teva has not shown that these finding are so clearly in error that this Court should have “a definite and firm conviction that a mistake has been committed,” the requisite for reversal. And Teva forgets that inequitable conduct cannot be found solely on materiality where there is no evidence of intent to deceive the PTO. Instead, Teva essentially asks this Court to reweigh de novo the evidence and draw a result opposite from the District Court.

One Response to “In Brief: Takeda v. Teva”

1. RSK
   November 6th, 2008 @ 02:14

   Could you please update me with what happened at the Court hearing?
   I dont see any major news company’ sites that have tracked this issue.
   Thank you.
   Regards,
   RSK

Leave a Reply

Name (required)

Email Address(required)

Website

Comments