by Dennis Crouch

Daikin Industries v. Chemours Co. (Supreme Court 2022).

Daikin’s IPR was successful, and the PTAB concluded that challenged claims of the Chemours patents were obvious. US Patent Nos. 7,122,609 and 8,076,431.  On appeal though, the Federal Circuit reversed — holding that the prior art “teaches away from the claimed invention” and that the Board also relied on non-analogous prior art.  Now Daikin has taken its case to the Supreme Court with a really smart petition focusing on the procedural divide between law and fact in the administrative law context.  The basic argument here is that the Federal Circuit grounded its decision on factual questions never expressly considered by the agency (the PTAB); but the proper procedure in that situation is a remand for agency determination of the facts rather than reversal.

Question presented:

May a federal appellate court, consistent with the Administrative Procedure Act and the principles of separation of powers embedded within that Act, reverse an administrative agency’s decision on a factual ground not addressed by the agency, without a remand to that agency?

Petition.  The petition here is supported by Judge Dyk’s dissent in the case where he argued that the majority opinion was “nothing less than appellate factfinding.”

The petition here goes on argue that the court’s approach here is part of the Federal Circuit’s modus operandi.

The Federal Circuit’s brand of aggressive judicial review, which casts aside basic principles of deference to administrative adjudication, was on full display in this case. Regrettably, it has long been embedded in the Federal Circuit’s practices, and its roots go even farther back than that. . . . This absence of deference to factfinders, even expert agencies, has been a hallmark of Federal Circuit review since that court’s creation. . . . To this day, the Federal Circuit remains isolated from the mainstream of administrative law. Scholars have noted that court’s “exceptional control” over the patent agency, and have labeled this phenomenon with terms such as “patent exceptionalism” and “Federal Circuit exceptionalism.” See, e.g., Peter Lee, The Supreme Assimilation of Patent Law, 114 Mich. L. Rev. 1413, 1415, 1452 (2016); Robin Feldman, Ending Patent Exceptionalism and Structuring the Rule of Reason: The Supreme Court Opens the Door for Both, 15 Minn. J.L. Sci. & Tech. 61 (2014); Paul R. Gugliuzza, The Federal Circuit as a Federal Court, 54 Wm. Mary L. Rev. 1791, 1818 (2013) (“The Federal Circuit gives minimal deference to PTO fact-finding.”).

There are a number of Supreme Court cases on point, but the key quote is below:

It is a ‘foundational principle of administrative law’ that judicial review of agency action is limited to ‘the grounds that the agency invoked when it took the action.’”

Dep’t Homeland Sec. v. Regents of Univ. of Cal., 140 S. Ct. 1891 (2020) (quoting Michigan v. EPA, 576 U.S. 743, 758 (2015)).

by Dennis Crouch

In Feb 2022, the Court of Appeals for the Federal Circuit released two separate opinions discussing the estoppel provisions of 35 U.S.C. § 315(e).

• California Inst. of Tech. v. Broadcom Ltd., — F.4th —, 2022 WL 333669 (Fed. Cir. Feb. 4, 2022)
• Intuitive Surgical, Inc. v. Ethicon LLC, — F.4th —, 2022 WL 414252 (Fed. Cir. Feb. 11, 2022)

Both decisions found that estoppel applied to bar certain post-IPR patent challenges.  That said, the two opinions are also in a amount of tension with regard to what claim challenges are estopped.  The decisions are important; Estoppel is a huge consideration of accused infringers as they consider how to structure their inter partes review (IPR) petition(s) and these are all multi-million-dollar cases.

Lets dig into the statute first, and then the decisions. The scope of estoppel was a major lobbying point during the passage of the America Invents Act of 2011 (AIA). Patentees raised concerns of harassment from multiple IPRs or IPRs followed by district court litigation. At the same time, Challengers were concerned that broad estoppel would leave them with no defense in court.

Under the 2011 statute, a patentee can challenge one or more claims of a patent via IPR. If an IPR is instituted, the IPR then typically ends with a final written decision determining whether to cancel the claims. Section 315(e) estoppel is triggered by that final written decision.  Once the final written decision issues, the IPR petitioner (or RPI/Privy) cannot do either of the following:

• [R]equest or maintain a proceeding before the Office with respect to that claim on any ground that the petitioner raised or reasonably could have raised during that inter partes review.  315(e)(1).
• [A]ssert either in a civil action arising [under the Patent Act] or in a [Section 337] proceeding before the International Trade Commission . . . that the claim is invalid on any ground that the petitioner raised or reasonably could have raised during that inter partes review. 315(e)(2).

You can see that these two subsections of 315(e) are substantially parallel and with only minor differences.  One commonality is that both tie estoppel to “the claim” or “that claim” — meaning the claim challenged by the petitioner.

In CalTech, Apple had filed several IPR petitions against CalTech’s patents.  Those were eventually won by the patentee.  Meanwhile, the infringement litigation was ongoing and Apple and Broadcom (the RPI) had raised several invalidity defenses.  However, the district court barred those defenses because they “reasonably could have [been] raised” in the IPR petition.  On appeal, the Federal Circuit affirmed that finding — particularly rejected the notion the “reasonably-could-have” expansion only applied to additional arguments that could have been raised post-institution.

[W]e take this opportunity to … clarify that estoppel applies not just to claims and grounds asserted in the petition and instituted for consideration by the Board, but to all claims and grounds not in the IPR but which reasonably could have been included in the petition.

CalTech.  As you read this quote from CalTech, focus in on the portion of the statement applying to claims: “estoppel applies . . . to all claims and grounds . . .  which could have been included in the petition.”  The implication from this statement in CalTech is that an IPR that challenges one claim of a patent could create estoppel vis-a-vis all claims of the patent.  Now, this could be a sensible rule, but it doesn’t appear to be the rule codified by the statute.

The CalTech estoppel was being battled-out in district court infringement litigation and so Section 315(e)(2) was the focus.  The second case, Intuitive Surgical looks instead to the parallel PTO-focused provision of Section 315(e)(1).

Intuitive filed three IPR petitions against the claims of Ethicon’s U.S. Patent No. 8,479,969.  All three petitions were granted and the PTO proceeded toward final written decision. But along the way the timing of the three became a bit offset.  In January 2020, the PTO issued final written decisions in two of the IPRs — both siding with the patentee and concluding the challenged claims had not been proven unpatentable.  At that point, the estoppel of Section 315(e)(1) kicked in — Intuitive could no longer maintain an IPR against , and the PTAB ruled that Intuitive could no longer maintain an IPR against the claim already blessed as patentable.  On appeal, the Federal Circuit affirmed that judgment over some fairly weak arguments from Ethicon. (Ethicon argued, inter alia, it was too tough and thus not reasonable to put all the arguments in one 14,000 word IPR petition. Ever heard of choosing your best arguments?).

In its opinion, the Intuitive court suggested that an IPR petition might avoid the estoppel by filing “multiple petitions where each petition focuses on a separate, manageable subset of the claims to be challenged—as opposed to subsets of grounds.”  According to the court, this approach avoids estoppel because “§ 315(e)(1) estoppel applies on a claim-by-claim basis.”  This conclusion appears directly contrary to the statement issued by the CalTech court one week prior.

Prior to CalTech, a the Federal Circuit had already ruled that estoppel applied on a claim-by-claim basis. See, for example, Credit Acceptance Corp. v. Westlake Services, 859 F.3d 1044, 1052 (Fed. Cir. 2017). However, those cases were based upon pre-SAS understanding of IPR procedure the source of those determinations appears to be Shaw Industries Group, Inc. v. Automated Creel Systems, Inc., 817 F.3d 1293 (Fed. Cir. 2016).  The Shaw decision is important here because it was expressly overruled and rejected by CalTech.  All this give the CalTech approach some credential.  However, I believe that ultimately we’ll see the “all claims … which reasonably could have been included in the petition” to be an errant bit of dicta that will not be followed by the subsequent panels.  I call it dicta, because the patentee in CalTech requested and obtained estoppel only for claims that Apple actually challenged via IPR petition.  There was no decision regarding non-petitioned claims.

All that said, the CalTech panel may want to consider rewriting this critical sentence to make clear that estoppel applies only to claim challenged in the petition, but to any ground that reasonably could have been included in the petition.

Accused patent infringers are fighting to overturn NHK-Fintiv rules that permit the PTAB to deny IPR institution in favor of pending litigation between the parties.

In addition to pending SCOTUS petitions; see a recent letter to the Commerce Sec. https://t.co/DszZkr2RG9 pic.twitter.com/pqOgtrtn73

— Prof. Dennis Crouch ⚖️ Patent Law & AI (@patentlyo) February 15, 2022

by Dennis Crouch

GlaxoSmithKline LLC v. Teva Pharmaceuticals USA, Inc., 18-1976 (Fed. Cir. 2022)

A divided Federal Circuit has denied Teva’s petition for en banc rehearing on the question of whether its FDA approved “skinny label” induced off-label infringement. The en banc rehearing vote appears to be 7-3  6-4 against rehearing. Chief Judge Moore and Judge Newman were both on the original panel. They were joined by Judges O’Malley, Taranto, Chen, and Stoll in denying the en banc petition.  Judges Prost was a member of the original panel and penned another dissent.  Judges Dyk, and Reyna each also wrote in dissent.  Judge Hughes’ vote was not announced, but I assume here that he voted in dissent with the majority.* Judges Lourie and Cunningham did not participate in the voting.

Although Teva lost its petition for rehearing, it really won the battle.  All of the judges agree that Teva “played by the rules” while GSK did something wrong.  On remand, Teva will apparently have the opportunity to argue this fairness issue under the doctrine of equitable estoppel.

• Statement that Patent Covers only CHF Use: The drug at issue here – carvedilol – is not patented. However, GSK has patented a particular use of carvedilol to “decrease a risk of mortality caused by congestive heart failure [CHF].” And, GSK provided a sworn statement to the FDA that identified CHF was the only patented use.
• Skinny Label: The FDA approved Teva’s sales of carvedilol, but only for limited uses under a “skinny label.”  In particular, Teva’s label says that its drug is for use only for treatment of hypertension and left ventricular dysfunction following a heart attack (“post-MI LVD”).  This skinny label approach carves-out the still patented use and is part of the Hatch-Waxman approach.   During Teva’s approval process GSK did not complain  or suggest that the skinny label uses would be infringing.
• Later in litigation, GSK argued that post-MI LVD uses statement induced infringement.

The original jury trial sided with the patentee, however Judge Stark flipped the verdict and issued a Judgment as a Matter of Law for the Teva. On appeal, the Federal Circuit flipped again and majority sided with the patentee — finding that the label coupled with other actions by Teva showed culpable intent to induce infringement of GSK’s use patent. However, here the majority walks through its explanation of Teva’s defense is equitable estoppel – rather than non-infringement.  And, the district court left that issue for decision on remand.

Equitable estoppel, a doctrine designed to avoid injustice, has three elements: misleading conduct, reliance, and prejudice.

In the patent context, the typical misleading conduct must someway reasonably inform the alleged infringer that the patentee does not intend to take action against certain conduct.   The majority here explains that estoppel is different from infringement analysis because estoppel includes a major focus on the patentee’s conduct rather than the accused infringing acts.

The dissent argues that the equitable estoppel approach is likely inadequate and unduly benefits the patentee.  Rather, an FDA approved skinny label that carves out all uses claimed-to-be-patented should absolve a generic maker from any inducement liability based upon the the label. “How could this label, which faithfully followed what the brand said about its own patents and which the FDA required Teva to use, itself be evidence that Teva intentionally encouraged something it knew would infringe? . . . When a generic plays by the skinny-label rules, the FDA-required label can’t be evidence of intent.” Prost in dissent.

The dissents generally argue that that the majority approach here gives branded companies an incentive to mislead the FDA and generic competitors in order to later catch them for infringement.

= = =

The opinion does not indicate how Judge Hughes voted, but Prof. Karshtedt and another reader suggested that we can infer that he voted with the majority.  The Federal Circuit’s internal operating procedure indicate that the clerk’s opinion should note any dissenting votes. And here, Judge Hughes vote was not noted as dissenting.

President Biden’s nominee to replace Judge Kathleen O’Malley has been confirmed by the U.S. Senate. Congratulations Judge Leonard Stark, and welcome to the Court of Appeals for the Federal Circuit.  O’Malley’s announced retirement date is March 11, 2022.  Assuming that has not changed, I expect that Judge Stark will be sworn in that same day or shortly after.  Judge Stark has handled more than 2,000 patent cases, including dozens of trials during his time as a Federal District Court judge in Delaware.

by Dennis Crouch

PersonalWeb has added a second petition for certiorari to the Supreme Court docket.  The first case focuses on res judicata and the Federal Circuit’s peculiar Kessler doctrine. That case has seen some action and is awaiting the views of the Solicitor General. See Resorbing Patent Law’s Kessler Cat into the General Law of Preclusion.

The new petition focuses on eligibility and asks the Supreme Court to reaffirm two separate pathways for computer-implemented business method inventions:

1. Improving “the functioning of the computer itself;” and/or
2. Causing an improvement in “any other technology or technical field.”

PersonalWeb argues that the Federal Circuit has focused only on the first prong, and only in a limited sense.   [Petition].

In Alice Corp., the Supreme Court used these terms in its Step 2 (something-significantly-more) analysis:

The method claims do not, for example, purport to improve the functioning of the computer itself.  Nor do they effect an improvement in any other technology or technical field. Instead, the claims at issue amount to “nothing significantly more” than  an instruction to apply the abstract idea of intermediated settlement using some unspecified, generic computer.

Alice Corp. Pty. Ltd. v. CLS Bank Int’l, 573 U.S. 208 (2014) (quoting Mayo).  During the prior appeal, PersonalWeb argued that its “True Name” file structuring system offered significant improvements to the technical field of computer network operations, but that argument was disregarded by the Federal Circuit who concluded that the invention was a “pure” abstract idea.

= = =

The patents at issue here are part of the patentee’s “True Name” family. U.S. Pat. Nos. 6,415,280; 7,802,310; and 7,949,662.  The patent begins with a description of traditional haphazard computer file storage systems where each user picks a different name and location for files.  This works fairly well for well-memoried individuals, but gets a bit crazy with distributed multi-user systems.  “[U]sers could save the same file with different filenames, or different files with the same filename, leading to unnecessary duplication, and confusion as to which files are actually in the system.”  The inventors here apparently learned of recent “fingerprinting” algorithms that were being used for public-key encryption and then had the idea that the same technology could be used for creating unique file identifiers for computer file systems — i.e., “true names.”  But truthfully, appears to be just a hash.  The asserted claims use the true names to control file access; retrieve files based upon their content; and remove duplicate files on the system.   The claims require use of a “content-dependent name”:

[T]he content-dependent name being based, at least in part, on at least a function of the data in the particular data item, wherein the data used by the function to determine the content-dependent name comprises at least some of the contents of the particular data item, wherein the function that was used comprises a message digest function or a hash function, and wherein two identical data items will have the same content-dependent name.

‘310 Patent, Claim 24.   Claim 24 is a retrieval function that retrieves a file using the content-dependent name.  The claim also includes an authorization check (such as a copyright license check) and will not permit access when “not authorized.”  The patentee here explains how the security check is facilitated by the True Name system.

The technical improvement that this invention provided over conventional file systems is clear. In conventional systems—where files are identified by user-selected file
names—a user could easily circumvent this authorization check simply by changing the file name. By using True Names, however—which always uniquely identify each file—the system ensures that only authorized users can access the file.

Petition.   The Federal Circuit also heard these arguments, but concluded that everything there “is all abstract.”

A growing trend pushes IP disputes arising from any agreement into arbitration or a particular court (and thus out of the PTAB).

Question: Should Congress pass a law prohibiting agreements not-to-challenge patents via inter partes review? (making such agreements unenforceable)

— Prof. Dennis Crouch ⚖️ Patent Law & AI (@patentlyo) February 9, 2022

by Dennis Crouch

For several years we have been tossing around the question of whether no-IPR contracts are enforceable. In Nippon Shinyaku v. Sarepta Therapeutics (Fed. Cir. 2022), the court says “YES THEY ARE” (at least when an alternate forum is provided).  Although the court did not enter into any serious policy analysis or consideration of Supreme Court precedent  promoting patent challenges such as Lear, Inc. v. Adkins, 395 U.S. 653 (1969); & MedImmune, Inc. v. Genentech, Inc., 549 U.S. 118, 124 (2007). (more…)

by Dennis Crouch

On Feb 4, the Federal Circuit issued an important decision in California Institute of Technology v. Broadcom Ltd. and Apple Inc. (Fed. Cir. 2022).   Four of the big issues are listed below:

• IPR Estoppel: The most important legal precedent in the case is a pro-patentee interpretation of the IPR estoppel provision found in 35 U.S.C. § 315(e)(2).  That prevision bars an IPR petitioner involved in patent litigation from asserting any invalidity grounds that the petitioner had “raised or reasonably could have raised during that inter partes review.” Id.  Here, the court makes clear that the scope estoppel is not tied to the grounds raised to in the IPR, but rather truly does extend to any ground “reasonably could have been included in the petition.”  Slip Op.  This decision expressly overrules Federal Circuit precedent to the contrary. Shaw Industries Group, Inc. v. Automated Creel Systems, Inc., 817 F.3d 1293 (Fed. Cir. 2016).
• Big money: The jury had awarded a $1 billion+ damage award; the Federal Circuit vacated that award and ordered a new new trial on damages.  The court also vacated an infringement finding with regard to one of the patent claims because the district court did not instruct the jury on the construction of a particular term (“variable number of subsets”).
• Eligibility: The inventive feature here is use of bit repetition to better ensure wireless signal transmission.  On appeal, Broadcom argued that the claim lacked eligibility because it employs a mathematical formula.  The Federal Circuit dismissed that argument out-of-hand with a direct quote from the Supreme Court. Diamond v. Diehr, 450 U.S. 175, 187 (1981) (“[A] claim drawn to subject matter otherwise statutory does not become  nonstatutory simply because it uses a mathematical formula, computer program, or digital computer.”).
• Extraterritoriality: The infringing chips are made abroad and installed in Apple devices abroad.  U.S. patents only cover infringement within the  territorial bounds of the United States.  The jury found infringing sales in the U.S.  On appeal, the Federal Circuit found the jury instructions appropriate.  Notably, the court refused to require any jury instructions  on the presumption against extraterritorial application of U.S. patent laws.  Rather, the instructions properly walked through factors for determining whether a particular sale occurred in the United States.

= = = =

In this post, I am just going to focus on the estoppel issues: Argument estoppel is a big deal with inter partes review.  By design, patent challengers get one-bite at the Apple; one shot at invalidating the patent claims based upon obviousness or anticipation.  Of course, the devil is in the details with this sort of rule.

Rather than relying upon traditional judge-made principles of res judicata, Congress specified within the statute how estoppel works for IPR proceedings.  The basic rule is that, once a patent claim is subject to a final-written-decision in an IPR, the IPR petitioner is estopped from asserting “that claim is invalid on any ground that the petitioner raised or reasonably could have raised during that inter partes review.” 35 U.S.C. 315(e)(2) (applies also to real-party-in-interest and privies).  Timing here is important, but favors estoppel.  Namely, once the IPR reaches Final Written Decision, the challenger is prohibited from continuing to assert invalidity, even if the litigation was already filed and had been pending prior to IPR institution.

The scope of estoppel provided by 315(e) has been subject to substantial litigation. One key decision is Shaw Industries Group, Inc. v. Automated Creel
Systems, Inc., 817 F.3d 1293 (Fed. Cir. 2016).  In Shaw, the Federal Circuit substantially limited the scope of estoppel. Shaw was a partial institution case — the PTAB had instituted IPR on only some grounds.  The court particularly held that 315(e) did not estopp the petitioner from later raising the non-instituted challenges in district court litigation.  The court reasoned that those could not  have reasonably been raised in the IPR since the petitioner’s attempt had been denied by the PTAB.   But Shaw raised further questions about where to draw the line, and district courts across the country came-up with a variety of conclusions about the scope of estoppel.  The most pro-challenger readings focused on grounds that could have been raised after institution, and thus concluded that estoppel was fairly strictly limited only to the grounds actually instituted.  See, e.g., Koninklijke Philips N.V. v. Wangs All. Corp., 2018 WL 283893, at *4 (D. Mass. Jan. 2, 2018).

Shaw was based upon a procedural posture that the Supreme Court eventually ruled improper.  Notably, Shaw presumed that partial IPR institution was proper. In SAS, the Supreme Court rejected that approach and instead held that IPR institution is an all-or-nothing decision by the USPTO. SAS Institute, Inc. v. Iancu, 138 S. Ct. 1348 (2018) (PTAB does not have partial institution authority; and that the IPR petition defines the scope of the IPR).

Ordinarily, each Federal Circuit panel is bound to follow precedent set-out by prior a Federal Circuit panel.  My colleague Tommy Bennett refers to this as the “Rule of Circuit Precedent.”  However, since Shaw‘s foundation had been undermined, the panel here in CalTech concluded that it was no longer binding precedent.

The panel here has the authority to overrule Shaw in light of SAS, without en banc action.

Although SAS did not expressly overrule Shaw, the court concluded that the Supreme Court had “undercut” Shaw‘s “theory [and] reasoning . . . in such a way that the cases are clearly irreconcilable.”  Quoting Henry J. Dickman, Conflicts of Precedent, 106 Va. L. Rev. 1345 (2020).

With this newfound freedom, the court rewrote the rule of estoppel — this time in a much broader sense:

Accordingly, we take this opportunity to overrule Shaw and clarify that estoppel applies not just to claims and grounds asserted in the petition and instituted for consideration by the Board, but to all claims and grounds not in the IPR but which reasonably could have been included in the petition.

Slip Op. In the litigation, Apple and Broadcom had hoped to rely upon several prior art references during litigation.  However, evidence showed that they were “aware” of those references by the time that they filed the IPR petition.  Even though they were not included within the IPR petition, they “reasonably could have” been used as the basis for an IPR challenge. Thus, the appellate panel affirmed the district court’s ruling barring the subsequent challenge in litigation.

The “aware of” requirement is quite broad, and the Federal Circuit did not delve further into defining the “reasonably could have” threshold. For its part, the district court in this case had gone a step further and expanded estoppel to also include any grounds relying upon prior art “a skilled searcher conducting a diligent search reasonably could have been expected to discover.” California Inst. of Tech. v. Broadcom Ltd., 2018 WL 7456042 (C.D. Cal. Dec. 28, 2018) (subsequent modifications omitted). This test is a based upon a statement made by Senator Kyl as the America Invents Act was being debated. 157 Cong. Rec. S1375 (daily ed. Mar. 8, 2011).  In his testimony, Sen. Kyl expressed that the estoppel provision had been “softened” by adding “reasonably.”  Without that term, the estoppel would have been

amenable to the interpretation that litigants are estopped from raising any issue that it would have been physically possible to raise in the inter partes reexamination, even if only a scorched-earth search around the world would have uncovered the prior art in question. Adding the modifier ‘‘reasonably’’ ensures that could-have-raised estoppel extends only to that prior art which a skilled searcher conducting a diligent search reasonably could have been expected to discover.

Statement of Sen. Kyl.

Although it is unclear how far reasonableness extends, it is clear that IPR estoppel is now based upon a use-it-or-lose-it principle.  Holding arguments back at the IPR petition stage is effectively a waiver of those arguments.