A panel of the Federal Circuit recently sounded off with something less than the usual deferential tone regarding review under the substantial evidence standard of factual findings on patentability made by the Patent Trial and Appeal Board (PTAB) in inter partes review (IPR) proceedings. This may not be surprising, however, because two of the panel judges were members of a four-judge group who previously questioned or, in the case of Judge Newman, outright rejected use of the substantial evidence standard for review of the PTAB’s factual findings in IPRs, preferring instead a more exacting and less deferential review of PTAB factfinding.

In its recent decision in OSI Pharm., LLC v. Apotex, Inc.,1 the Federal Circuit reversed a finding of obviousness of OSI”s patent claims directed to the drug erlotinib (sold as Tarceva) for treating non-small cell lung cancer (NSCLC). While the opinion presents a methodical reversal of the PTAB’s finding that there was substantial evidence of a reasonable expectation of success in the prior art, the Court cloaked its reversal in strong language from the Supreme Court’s opinion in Dickinson v. Zurko2 relating to the standard of review for agency factfinding. Emphasizing “the importance of not simply rubber-stamping agency factfinding,” the Federal Circuit further quoted the Supreme Court’s pronouncement that the Administrative Procedural Act requires “meaningful review” and “stricter judicial review of agency factfinding than Congress believed some Court’s had [conducted before the Act].”3 This may raise some eyebrows as to whether the decision signals a change at the Federal Circuit to stricter review of the PTAB’s findings under the substantial evidence standard, with an eye toward more reversals, or whether the decision is limited to a rehash of the views of particular judges who have previously questioned whether the substantial evidence standard should even be applied in the first place.

At the time of Dickinson in 1999, the primary mechanism for post-grant review of patent validity was ex parte reexamination by the Patent and Trademark Office (PTO). Up to that time, the standard of appellate review of the agency’s factual findings in those proceedings had been the clearly erroneous standard that likewise applies to factual findings in district court proceedings, which the Court referred to as the “court/court standard.”4 In Dickinson, the PTO commissioner convinced the Supreme Court to overturn the en banc Federal Circuit’s embrace of the clearly erroneous standard and hold that the standard of review for agency decisions should be those set forth in the Administrative Procedure Act.5 The Court noted that there could be “some practical difference in outcome depending upon which standard is used,” but the Court seemed to shrug off the difference as a “subtle one” that was “so fine that (apart from [that] case) we have failed to uncover a single instance in which a reviewing court conceded that use of one standard rather than the other would in fact have produced a different outcome.”6 The Court concluded that the “court/agency standard” should apply, which the Court recognized as being “somewhat less strict than the court/court standard.”7

The Federal Circuit subsequently applied Dickinson in In re Gartside8 and chose the “substantial evidence” standard rather than the “arbitrary and capricious” standard in the APA. The Federal Circuit, which had previously sought en banc to hang on to the clearly erroneous standard9 labeled the arbitrary and capricious standard as “highly deferential” and chose the “less deferential review standard” that “asks whether a reasonable fact finder could have arrived at the agency’s decision.”10

Application of the substantial evidence review standard for agency patentability determinations proceeded without fanfare for more than a decade, including the transition to IPR proceedings at the PTAB initiated by the America Invents Act, until 2016 when a petition for a rehearing en banc was decided in Merck & Cie v. Gnosis S.P.A.11 Patentee Merck asked the Federal Circuit to decide whether the clear error standard of review should be required for appeals of factfindings in IPR proceedings. The issue got enough attention from four judges who offered opinions on the court’s ultimate denial of the petition.

Judge O’Malley wrote an our-hands-are-tied concurring opinion that was joined by Judges Wallach and Stoll. They felt that the substantial evidence standard is “seemingly inconsistent with the purpose and content of the AIA” and “makes little sense” in the context of an IPR appeal, but they concluded they were bound to continue to apply it by the Supreme Court’s precedent in Dickinson and their own in Gartside.12 Despite making their case for a switch to the clear error standard for IPR review, the judges concluded that such a switch would have to be made by Congress or the Supreme Court, not the Federal Circuit. The judges grounded their opinion in the court/court versus court/agency distinction that was the foundation of Dickinson, and distinguished ex parte reexamination proceedings from IPR proceedings to the extent IPRs were “intended to replace district court litigation.”13 They believed that the court/court clear error review standard for IPRs would follow if “Congress meant to create an adversary, party-instituted proceeding to consider what would otherwise be considered by district courts.”14 Given the number of instances where district court litigation is routinely stayed in view of IPR proceedings, it seems unlikely the other Federal Circuit judges would disagree with their premise. However, in the absence of statutory language from Congress in the AIA addressing the standard of review, they concluded that “the fact that it might have made sense for Congress to change [the] standard of review for IPRs does not mean it did so.”15

Judge Newman dissented in Merck and saw an avenue for the Federal Circuit to apply the clear error standard for factfinding in IPR proceedings. To get over the precedent hurdle cited by the concurring judges, Judge Newman saw that Congress created a “new tribunal with authority to substitute for district court actions” such that “Congress expected that these PTAB decisions would be reviewed on the same judicial standard.”16 Judge Newman seemed to equate the substantial evidence standard with a “rubber-stamp” where application of it can lead to affirmance of a ruling inconsistent with the weight of the evidence and concluded that it is “unlikely that Congress intended to place PTAB decisions in that ‘rubber-stamp’ category.”17 She provided an extensive analysis of the legislative record for the AIA and concluded that it “provides the expectation the Federal Circuit will apply the standard judicial criteria for review.”18

This set the stage for the recent OSI decision. The language of the opinion arguably reflects an if-you-can’t-beat-them-join-them embrace of the substantial evidence standard to nevertheless get to the place where some of the judges wanted to go in Merck, namely less deferential review of the PTAB’s factual findings. Judge Stoll wrote the opinion and she and Judge Newman were joined on the panel by Judge Taranto. Judge Stoll appeared to take a close look back at Dickinson and quoted strong language from it in what might reflect an attempt to sharpen the teeth of a substantial evidence review of the Board’s factfindings. In Merck, Judge Newman raised her “concern” with the “highly deferential” substantial evidence standard and distinguished it from a “full and fair review” that she deemed appropriate for IPR appeals. Now in OSI, however, the Federal Circuit layered the “meaningful review” and “stricter judicial review” language from Dickinson onto the substantial evidence standard and used it to reverse the PTAB. Again going back to Merck, Judge Newman seemed to place the substantial evidence standard in the “rubber-stamp category.” Now in OSI, however, the Federal Circuit appeared to take it out of the category and wrapped the Supreme Court’s warning on the “importance of not simply rubber-stamping agency factfinding” around the substantial evidence review analysis. It is notable that Judge Taranto joined in this approach to IPR review in OSI even though he did not join any of the writers in Merck. It is clear from reading the court’s dissection of the evidence relied upon by the PTAB in OSI and its takedown of the Board’s findings that the panel engaged in anything but a rubber-stamp analysis.

Only time will tell if other Federal Circuit judges take the approach of the OSI panel to lead to further reversals of the PTAB. Given the number of decisions on IPRs coming out of the Federal Circuit on an almost daily basis, it may not take long to find out.