On January 20, 2015, the Supreme Court issued its long-awaited decision on the standard of review of factual findings by the trial court in construing patent claims.  The Court ruled that factual findings in the context of claim construction are subject to Rule 52(a)(6) of the Federal Rules of Civil Procedure, which sets out a deferential “clearly erroneous” standard of review on appeal. The Court thus established two standards of review for claim construction rulings: “clearly erroneous” review of factual findings based on extrinsic evidence, and de novo review for legal findings, including the meaning of the intrinsic evidence (the claims, specification, and file wrapper) and “the ultimate question of proper construction of the patent.”  Teva Pharmaceuticals USA, Inc., et al. v. Sandoz, Inc., et al., 574 U.S. (2015).

The Supreme Court granted certiorari in the Teva Pharmaceuticals case to review the Federal Circuit’s rule that findings of fact in claim construction are subject to the stricter de novo review standard. Teva Pharmaceuticals had sued Sandoz Inc. and others for infringement after they applied for pre- market approval from the FDA of generic versions of a Teva-produced drug for treating multiple sclerosis.  Sandoz argued to the trial court that a particular claim term – “molecular weight” – was subject to three different interpretations and thus was invalid for indefiniteness. Both parties presented expert evidence regarding what an ordinary artisan would have understood “molecular weight” to mean at the time of the invention.  The District Court found Teva’s expert more compelling, and rejected Sandoz’s indefiniteness argument. On appeal, the Federal Circuit, applying the de novo standard of review, determined that Sandoz’s experts were more credible, and that the “molecular weight” term was indefinite because it was subject to multiple interpretations. Teva applied for a writ of certiorari on the basis that the Federal Circuit applied the wrong standard for reviewing the trial court’s factual findings.

In a 7-2 opinion, the Supreme Court held that “clear error” is the appropriate standard of review for factual findings underlying a district court’s claim construction ruling. The Court clarified that it was not departing from its opinion in Markman—claim construction, like contract interpretation, is a matter of law, and therefore subject to de novo review on appeal.  However, the Court explained, analogizing again to the interpretation of contracts or deeds, sometimes, such as when technical terms are used, a factual dispute arises.  When that happens in claims construction, resort to extrinsic evidence such as expert testimony may be needed. Determinations of that extrinsic evidence, the Court held, are factual in nature and fall under Rule 52(a)(6) of the Federal Rules of Civil Procedure.

Rule 52(a)(6) specifies that questions of fact will be reviewed on appeal under the “clearly erroneous” standard.  The Court explained that Rule 52 does not provide for any exceptions that would permit a different standard of review for factual findings in the claims construction phase of patent litigation. Further, the Court noted, this conclusion does not conflict with its precedent in Markman, because that case did not specify the standard of review for findings regarding extrinsic evidence. Finally, the Court cited precedent and practical considerations—including noting that trial judges are more exposed to the technology and evidence in each case and are generally better able to weigh factors such as the credibility of experts –as supporting the use of the “clearly erroneous” standard.

The Court rejected the arguments that conducting review under two standards would be difficult for the Federal Circuit and that it would be difficult to separate legal and factual issues. Citing its own as well as Federal Circuit and other Circuit precedents, the Court noted that appellate courts have a long history of applying disparate standards of review to factual and legal issues in patent litigation and in other contexts. The Court also noted that some of the Federal Circuit precedents create their own complexity (as they appear to blur the standards), citing, inter alia, Dow Jones & Co. v. Ablaise Ltd., 606 F. 3d 1338, 1344–1345 (Fed. Cir. 2010) (“[T]his court,” while reviewing claim construction without deference, “takes into account the views of the trial judge”); Nazomi Communications Inc., v. Arm Holdings, PLC, 403 F. 3d 1364, 1371 (Fed. Cir. 2005) (“[C]ommon sense dictates that the trial judge’s view will carry weight” (citation and internal quotation marks omitted)); and Lighting Ballast Control LLC v. Philips Electronics North Am. Corp., 744 F. 3d 1272, 1294 (Fed. Cir. 2014) (Lourie, J., concurring) (we should “rarely” overturn district courts’ true subsidiary fact- finding; “we should, and do, give proper informal deference to the work of judges of a subordinate tribunal”).

The other counterargument considered by the Court concerned possible lack of uniformity due to less strict Federal Circuit oversight of claim construction rulings. The Court found that this was of minimal concern because of the issue- preclusive nature of prior rulings and because prior opinions will at least provide persuasive authority. Also, the opinion repeats the finding in Markman that extrinsic evidence is “unlikely to loom large in the universe of litigated claim construction.”

The Court concluded its analysis by providing an example of how to apply this standard of review. The Court explained that, in instances where no extrinsic evidence is required to interpret a patent, where the judge is relying solely on the intrinsic evidence, the claim construction is a purely legal matter subject to de novo review. Where extrinsic evidence is considered as to one or more terms, the interpretation of the evidence constitutes a factual finding, which is subject to  the “clearly erroneous” standard under Rule 52(a)(6).

The Court applied this new approach to the facts of the instant case, and explained that the District Court was required  to weigh competing expert opinions about the interpretation of “molecular weight.”  This represented a factual finding based on extrinsic evidence, and because the Federal Circuit applied the de novo standard on review, the Supreme Court vacated the Federal Circuit decision and remanded the case to the Federal Circuit to apply the correct standard.

The Teva Pharmaceuticals decision marks a shift in appellate review of claim construction rulings, but only for those cases in which extrinsic evidence plays a role in claim construction. The Court correctly noted that (a) most constructions should not be implicated because most should turn on the intrinsic evidence; (b) trial courts are generally better positioned than appellate courts to weigh certain evidence, such as the credibility of experts; and (c) the Federal Circuit has jumped through some semantic hoops in claiming to apply the “de novo” standard while acknowledging at times the need to give credence to a trial judge’s factual findings.  Nevertheless, the revised standard of review will likely incentivize litigants to invoke extrinsic evidence more often in claims construction. If that happens, the Markman phase will become more involved and costly. Moreover, an increased reliance on extrinsic evidence may reduce the likelihood of summary judgment and may result in trial judges delaying the claim construction phase until after discovery or even trial, further increasing the costs of patent litigation.