In inter partes review proceedings, is the U.S. Patent and Trademark Office's Patent Trial and Appeal Board required to take into account a final district court determination of non-obviousness of the same claims based on the same prior art?  In Novartis AG v. Noven Pharmaceuticals Inc., the Federal Circuit responded in the negative.  Of course, this ruling will have an impact on the many parallel Patent Office and district court proceedings, which have been increasing significantly since the passage of the AIA.  Ultimately, the Court rested its decision on the different burdens of proof required in the two different types of proceedings.  Therefore, as a matter of law, "the PTAB properly may reach a different conclusion based on the same evidence."  The Federal Circuit defended its decision by citing to the Supreme Court's Cuozzo Speed Techs., LLC v. Lee case, which noted that the "different evidentiary burdens mean that the possibility of inconsistent results is inherent to Congress'[s] regulatory design."  Cuozzo, 136 S. Ct. 2131, 2146 (2016).  But is there a problem with a system in which a patent owner's rights can sit precariously between the "preponderance of the evidence" and the "clear and convincing evidence" burdens of proof.

To be fair, Novartis did not argue during the oral hearing that the Board should be bound by the District Court decision.  Instead, it took the position that the Board should be required in such circumstances to explain why the difference in burdens of proof matters:  "The Board should have recognized the findings of the district court and explained why it came to a different conclusion on the same evidence that was before the district court."  Oral Hearing at 3:48.  Such an argument makes sense in view of prior IPR appeals that resulted in the Federal Circuit vacating and remanding the Board's decision.  For example, appellants have been successful in convincing the Court that when the PTAB's final written decision does not sufficiently explain the rationale for invaliding a claim, it must be remanded.  Nevertheless, the panel was quick to point out that the Board in this case not only explained why the claims were obvious, but it explained that the difference in burdens of proof allowed it to come to a different conclusion.  In other words, the Federal Circuit held that the Board is not required to take the additional step to explain why the prior art might be invaliding under one standard, but might not be invalidating under the other.

This case is not significantly different than In re Baxter International, Inc., 678 F.3d 1357 (Fed. Cir. 2012).  In that case, the Patent Office invalidated claims in a reexamination that were previously upheld by a district court.  The difference between these two cases, other than the type of Patent Office proceeding involved, was that in Baxter the reexamination rejections were based "on prior art references that were not squarely at issue during the trial on the invalidity issues . . . ."  Nevertheless, the Federal Circuit came to the similar conclusion in both cases, that because "the two proceedings necessarily applied different burdens of proof and relied on different records, the PTO did not err in failing to provide the detailed explanation now sought by Baxter as to why the PTO came to a different determination than the court system in the Fresenius litigation." 

Novartis, however, focused on a subsequent comment found in the Baxter case:  "When a party who has lost in a court proceeding challenging a patent, from which no additional appeal is possible, provokes a reexamination in the PTO, using the same presentations and arguments, even with a more lenient standard of proof, the PTO ideally should not arrive at a different conclusion" (emphasis added).  It essentially argued that the use of the word "ideally" was a signal to the Patent Office that on identical records, the outcomes must be the same.  But Chief Judge Prost responded during the oral hearing:  "And let's hope it doesn't snow tonight."  In other words, as the opinion makes clear, the use of the term "ideally" in Baxter appeared to connote aspiration, not a mandate.  Moreover, again as Chief Judge Prost pointed out during the hearing, the Baxter decision follows up that above sentence with the all-important word:  "However . . . ."

Judge Wallach, who authored the Novartis AG opinion, had another observation that also dispenses with the "ideally" language from Baxter -- the factual records between the cases were not the same.  Sure, for at least one of the patents at issue, the prior art relied upon by the Board was exactly the same as the art cited by the District Court, and the same experts were used in both cases.  But as Judge Wallach pointed out, the expert reports were not identical in the two proceedings.  In fact, Novartis had to acknowledge that that the experts' testimony was only "substantively the same."  As the Court put it, "[i]t is unsurprising that different records may lead to different findings and conclusions."  But such an outcome appears to put form over substance -- experts will almost never prepare identical reports in different proceedings and the deposition testimony will certainly vary both in the questions being asked and the exact answer that are provided.  But the cited art does not change, and the issue of whether the claims are anticipated or are obvious should not fluctuate based on the length of the expert report or whether the expert was asked the same questions at her deposition.  Moreover, it sets up incentives to harass a patent owner by providing only slightly different testimony in the two different proceedings, thereby potentially getting two bites at the apple.

The technology at issue in this case was a pharmaceutical composition comprising both a compound known as rivastigmine and an antioxidant.  Such a composition is described as being useful for the treatment of Alzheimer's disease.  The Board had invalidated the relevant claims as obvious in view of two references:  Enz and Sasaki.  Enz disclosed a rivastigmine transdermal patch containing an acrylic polymer.  Sasaki was described as teaching that compounds combined with acrylic polymers tend to have their therapeutic effect reduced due to breakdown and dissipation of the drug over time; and that adding an antioxidant will slow this process.  With regard to its substantive arguments, Novartis focused on the motivation to combine.  Specifically, it alleged that there was no evidence that a PHOSITA would have been motived to add an antioxidant "absent evidence of oxidative degradation."  As counsel put it during the oral hearing, the old adage applies:  "if it ain't broke, don't fix it."

Nevertheless, the Federal Circuit found that there was substantial evidence for the Board's decision.  It first pointed out that the Board found that the level of skill of a PHOSITA included the ability to analyze and recognize certain characteristics of a compound.  Moreover, the Court reduced the remaining assertions of Novartis to requests to give more weight to Dr. Alexander M. Klibanov testimony, Novartis' expert.  And of course, the Court declined:  "Novartis asks us to give greater weight to the testimony of Dr. Klibanov than did the PTAB, which we may not do."  As a result, the Court affirmed the final written decisions of the Board related to the patents at issue.

Novartis AG v. Noven Pharmaceuticals Inc. (Fed. Cir. 2017)
Panel: Chief Judge Prost and Circuit Judges Wallach and Stoll
Opinion by Circuit Judge Wallach