In Regeneron Pharmaceuticals v. Merus N.V., Appeal No. 2016-1346 (Fed. Cir., July 27, 2017), a split three-judge panel of Prost, Wallach and Newman (Newman dissenting) affirmed the district court’s ruling that claim 1 of Regeneron’s U.S. Pat. No. 8,502,018 is unenforceable due to inequitable conduct (IC). The ruling was unusual since the court originally stated that it would hold two hearings; one on the materiality of 4 nondisclosed references, and another on the specific intent to deceive element of IC. Claim 1 reads:
“A genetically modified mouse, comprising in its germline human unrearranged variable region gene segments inserted at an endogenous mouse immunoglobulin locus.”
The district court judge began by construing the claim using the broadest reasonable interpretation, and rejected Regeneron’s arguments that claim 1 should be limited to “reverse chimeric” antibodies that contained human sequences only in part of the variable region, while leaving the remainder of the heavy and light chains 100% mouse sequences. Rather, the judge found that the use of “comprising” in the claim permitted the claim to encompass “humanized” antibodies, in which all but part of the variable region was “human” as well as fully human antibodies with no mouse protein. However, the Examiner’s statement of reasons for allowance simply stated that the prior art does not teach or suggest [the language of claim 1]. Regeneron’s arguments that the claim was limited to gene segments that would lead to “reverse chimeric” antibodies that contained human sequences only in part of the variable region, while leaving the remainder of the heavy and light chains 100% mouse sequences. Rather, the judge found that the use of “comprising” in the claim permitted the claim to encompass “humanized” antibodies, in which all but part of the variable region was “human” as well as fully human antibodies with no mouse protein. However, the Examiner’s statement of reasons for allowance simply stated that the prior art does not teach or suggest [the language of claim 1].
Not surprisingly, this broad reading of the scope of the claim led to a finding that the “withheld references” that Regeneron was aware of, but did not submit, were “but-for material” to the patentability of claim 1. Regeneron’s arguments that the references were cumulative was rejected.
The district court did not hold the bench trial on intent but went on to “sanction Regeneron by drawing an adverse inference of specific intent to deceive the PTO” due to Regeneron’s misconduct during litigation, including violating the court’s discovery orders and so the question central to the appeal, was whether or not litigation misconduct per se could meet the Therasense requirement for proof of specific intent by clear and convincing evidence. Specific intent can be inferred, but specific intent to deceive the PTO must be the single most reasonable inference that can be drawn from all of the evidence.
Therasense also established that IC can be based on egregious misconduct, which does not require but-for materiality of undisclosed references but rather, is based on misrepresentations to the PTO that led to the issuance of the patent in suit.
The Therasense court pointed out that the requirement for egregious misconduct was meant to align with the “unclean hands” defense to infringement applied by the Supreme Court in early cases such as “Precision, 324 US at 816-820 (perjury and suppression of evidence), Hazel-Atlas, 322 US 240 (manufacture and suppression of evidence) [and] Keystone, 290 US at 243 (bribery and suppression of evidence). The Therasense court stated that these were cases in which “the patentee acted knowingly and deliberately with the purpose of defrauding the PTO and the courts.” In each of these foundational decisions, the applicant obtained the asserted patent by fraud on the PTO, but related fraudulent acts continued during litigation. The penalty for this coming to the courts with unclean hands was dismissal of the suit, not the “sanction” of unenforceability.
The Therasence court addressed egregious misconduct by examples of prosecution misconduct. Thus, the question remained as to whether or not, in the court’s opinion, egregious misconduct during litigation could per se lead to a finding of the intent-to-deceive element of inequitable conduct and, coupled with but-for materiality of references omitted during prosecution, extinguish the enforceability of the patent.
In this case, inferring intent from Regeneron’s failure to submit but-for material references would not be a simple matter. What if Regeneron’s prosecutors in fact believed that the claims should be construed narrowly and that the omitted art – which was received after issuance of the notice of allowance—was not but-for material and/or was cumulative to the art of record? That the prosecution team should have known [better] and even gross negligence on its part are not sufficient evidence of specific intent to deceive.
The statements of one Dr. Smeland, the chief prosecutor, called problematic by the court, seem to support his belief after inquiry, that the omitted references were cumulative and that they did not teach the reverse chimeric “inventions” that the Examiner had allowed in the [application that lead to the] ‘287 patent.” The improper withholding of a chart and comments by Dr. Jones, another of the prosecutors, was considered to be evidence of litigation misconduct, but the opinion does not disclose what it revealed. Thirteen pages of the majority’s opinion are devoted to Regeneron’s “troubling misconduct” during discovery and trial.
It remains troubling to me and to Judge Newman, why the Aptix decision, Aptix Corp. v. Quikturn Design Systems, 269 F.3d 1369 (Fed. Cir. 2001) should not dictate the outcome of this suit. The majority concedes that, in Aptix, “we held that courts may not punish a party’s post-prosecution misconduct by declaring the patent unenforceable.” This makes sense, because the persons with a duty to disclose information during prosecution almost never have control over, or even knowledge of, the strategies employed by the litigation team, which is often not retained until years after the patent in suit was issued.
In this case, the majority attempts to finesse this argument by asserting that Regeneron was accused “not only of engaging of post-prosecution misconduct but also of engaging in inequitable conduct during prosecution…[and] Regeneron’s litigation misconduct, however, obfuscated its prosecution misconduct.” While the majority found that “only after Merus proved the remaining elements [sic] of [IC] did the district court hold the patent unenforceable”, there is no finding based on a clear analysis of the evidence of the prosecution team’s intent to deceive the PTO.
If you have read this far, you can see that I find Judge Newman’s dissent convincing. In fact, since the entire opinion is 57 pages long, I recommend that you read the first pages of the majority opinion that explain the technology involved in the transgenic antibodies and then read the 19 page dissent. In both opinion, you can skip pages of description of the teaching of the omitted references, since the adoption of a broad claim construction obfuscates the relevance of any given reference. Judge Newman’s central point is that dismissal of the suit, rather than a holding of unenforceability of the patent is the proper remedy is the proper remedy for litigation misconduct. Now go back to the Keystone Driller decision and see if you agree.
This is a Catch-22 situation, where the failure to prove specific intent to deceive during prosecution due to litigation misconduct, permits the litigation misconduct to be substituted for proof of specific intent by the prosecutors.