Belcher v. Hospira – Inequitable Conduct is Still in Your Invalidity Toolbox

Ever since Therasense made it more difficult to plea and prove inequitable conduct (IC), I feel that practitioners and litigators have pushed this defense to patent infringement out of the front lines of infringement attacks. In fact, I know of only a handful of cases in the pharma space in which an inequitable conduct defense was successful.  In a sense, it has been replaced by pleas for enhanced damages, sanctions for misbehavior during litigation, and even anti-trust violations. This despite invalidations of a variety of patents post-Therasense in decisions such as GS Cleantech v. Adkins Energy, Appeal No. 2016-2231, 2017-1838 (Fed. Cir., March 2, 2020), Apotex v. UCV, 763 F.3d 354 (Fed. Cir. 2019) and Aventis Pharma. v. Hospira, 675 F.3d 1324 (Fed. Cir. 2012).

In Belcher Pharmaceuticals  v. Hospira, Appeal No. 2020-1799 (Fed. Cir., Sept. 1, 2021), a Fed. Cir. panel of Judges Reyna, Taranto and Stoll (Reyna writing) affirmed the district court’s finding that Belcher’s U.S. Pat. No.9,283,197 was unenforceable due to withholding material information from the PTO with the requisite deceptive intent. The  ‘197 patent claimed:

6. An injectable liquid pharmaceutical formulation of l – epinephrine (“L- E”) sterile solution…having a pH between 2.8 and 3.3; …compounded in an aqueous solution as 1.0-1.06 mg/ml [side-product impurity and stability data follow.]

These two parameters are important because, when an L-E solution becomes more acidic (i.e. pH decreases), racemization increases and oxidation deceases. Accordingly, quoting from a learned treatise the panel wrote that “there is a optimal pH at which racemization and oxidation can be balanced to minimize loss of [L-E] by these two routes; this is approximately pH 3.0-3.8.” Slip op. at 3-4. According to the NDA Belcher filed, the prior art disclosed 1.0 mg/ml concentration and a pH range of 2.8-3.3 (the reference product). Belcher’s pH was 2.4-2.6, A consultant recommended that Becher revert to pH 2.8-3.3 to avoid an FDA requirement to evaluate the effect of an in-process pH range of 2.8-3.3, and Belcher did so. This would prove to be bad advice.

In Belcher’s patent, “The inventors alleged discovery of raising the pH [from 2.2-2.6 to 2.8-3.2] …was a nonobvious solution to the problem of racemization.” In the first office action, the Examiner cited a Canadian patent application that disclosed a 1.0 mg/ml L-E solution that was preservative free, made in a nitrogen atmosphere and had a pH range of 2.2-5.0. The attorney cited in re: Woodruff for the proposition that, if a claimed range is critical and yields unexpected results,[reducing racemization in this case], it is nonobvious over an overlapping prior art range. The Examiner accepted this assertion and allowed the application.

This was a para. IV case. The parties agreed that there was no anticipation, so the arguments were doe infringement or inter alia unenforceability. A key Belcher witness was its “Chief Science Officer”—one Mr. Rubin, who was assigned all of the duties of a patent attorney—drafting prosecution and litigation—although he was neither an attorney nor a patent agent. He was responsible for obtaining the allowance of claim 6, above, which did not recite the use of any preservatives in the solution. He testified that he was aware that the reference product had a pH of  2.8-3.3 and he knew of a reference, Stepensky, that disclosed a pH 3.25-3.70 and that he had sent into the FDA. He also admitted that he was aware of a “JHP” L-E solution that had a 2.2-5.0 PH range. The district court ruled that the failure to cite these three references to the PTO met the Therasense standards for materiality and intent, and so the patent was unenforceable due to IC.

The Fed. Cir. panel stated that the standard for establishing but-for materiality “only requires a preponderance of the evidence, giving claims their broadest reasonable construction.” The panel cited Aventis v. Hospira repeatedly, first for the fact that the burden of showing materiality using the PTO’s evidentiary standards is lower than would be required for a district court to find a reference to be material using the “clear and convincing standard. The district court had found the claims obvious in view of the uncited art, and the panel felt that this was sufficient to establish the materiality of the uncited art. The panel ruled that Belcher could not now argue that the omitted references were cumulative, since they had argued that the claimed range was critical during prosecution, even though Belcher was aware that the range was in the public domain via the JHP product.

The district court had found, and the panel agreed, that, even without direct evidence of specific intent to deceive the Examiner, Rubin’s activities, particularly his “false” criticality arguments during prosecution, supported specific intent as the only reasonable inference. Belcher had argued that Rubin genuinely believed that the withheld references were irrelevant. The panel noted that the district court had not totally relied on the lack of credibility of Rubin’s testimony, but had relied on facts of record, including those relating to Rubin’s arguments to the PTO about the criticality of the range. The district court’s’ IC ruling was affirmed.

“I didn’t believe that a reference was relevant, so I didn’t cite it” sounds like a reasonable testimony from a company employee who is not an agent or an attorney, yet is in charge of making decisions regarding patent preparation and prosecution. In fact, IC has not been found when patent attorneys insufficiently evaluate the relevance of a piece of prior art. However, the moral of this case may be to get your facts—or the absence of facts—in order and offer consistent testimony. Here, Rubin was clearly a “man who knew too much.”

  Back to All Resources