IPRs and the APA: Review of Director’s Discretion to Initiate IPRs

By Jordan Duenckel.  Jordan is a second-year law student at the University of Missouri School of Law and a registered patent agent. 

Apple brought an action against the USPTO Director Vidal in district court under the Administrative Procedure Act (APA), 5 U.S.C. §§ 701– 706, challenging the Director’s instructions to the Board regarding exercise of discretion in IPR institution decisions. In Apple v. Vidal, 2022-1249, — F.4th — (Fed. Cir. Mar. 13, 2023), Judge Taranto (joined by Judges Lourie and Stoll) largely affirmed the district court’s dismissal, confirming that the Director’s instructions are unreviewable.  The court did separately reverse a tertiary challenge to allow Apple to proceed on a claim related to the note-and-comments procedure of the APA. 

Apple and other repeat players in patent infringement litigation often use the inter partes review process under 35 U.S.C. §§ 311–319 to challenge the validity of asserted patents. The statute provides a two-step IPR process: Step 1 is the institution decision by the Director under § 314(b); Step 2 is the trial and final written decision by the PTAB.   

At least two prerequisites assist the Director in deciding to grant review: [1] a reasonable likelihood that the petitioner would prevail in 35 U.S.C. § 314(a) and [2] a petition must be filed within one year after service of the infringement complaint. § 315(b). Even if these conditions are met, the Director has unreviewable discretion over whether to initiate an IPR. The statutory text is seemingly as clear as a statute can be: “The determination by the Director whether to institute an inter partes review under [§ 314] shall be final and non-appealable.” 35 U.S.C. § 314(d); see also United States v. Arthrex, Inc., 141 S. Ct. 1970, 1977 (2021).   

From the outset of the IPR program, the Director delegated institution authority to the Board. 37 C.F.R. § 42.4(a). Practically, without this delegation, Director Vidal would spend a disproportional amount of time reviewing IPR petitions at the expense of other duties of the office, although she could have delegated responsibility to other agency departments such as the petitions division.  The right of delegation of the institution is settled law. See Ethicon Endo-Surgery, Inc. v. Covidien LP, 812 F.3d 1023, 1031–32 (Fed. Cir. 2016).  

At issue in Apple v. Vidal are the so-called Fintiv instructions issued by the Director based on Apple Inc. v. Fintiv, Inc., IPR2020-00019, 2020 WL 2126495 (P.T.A.B. Mar. 20, 2020) which provides six factors for analysis of whether to institute an IPR parallel to pending litigation.   

Proposing an analysis under the arbitrary and capricious standard, Apple and the other petitioners are directly focused now not on the denial of a specific petition for IPR review but as a general challenge to the Director’s instructions to the PTAB about how to exercise the delegated discretion.   

Slip Op. The district court ruled that 35 U.S.C. § 314(d) “precludes judicial review” of the challenged agency actions, bringing the case within the APA exclusion stated in 5 U.S.C. § 701(a)(1). According to this court, the IPR statute’s preclusion was settled by the Supreme Court in Arthrex and encompasses review of content-focused challenges to the Fintiv instructions. § 314(d) provides the clearest congressional delegation of nonreviewable discretion possible and the panel rightfully relied on plain-meaning and clear Supreme Court precedent.    

While affirming the dismissal of the content-based claims, the court separates the procedural requirements set forth in the APA. Reversing the district court in part, Judge Taranto’s panel opinion reopened Apple’s claim that the Director was required, by 35 U.S.C. § 116 together with 5 U.S.C. § 553, to promulgate institution instructions through notice-and-comment rulemaking procedures.  Slicing the procedure from the underlying substance of the rule, Taranto relies on Lincoln v. Vigil, 508 U.S. 182, 195 (1993) to clarify that the 5 U.S.C. § 553 provides the basis for rulemaking through the notice-and-comment procedure for the Director’s instructions and is a separate analysis of reviewability from the substance of the instructions. 

Standing was also preemptively addressed for the remand proceedings. Lujan provides the three-step test: injury-in-fact, causation, and redressability. In search of a particularized, concrete injury, the court takes notice that Apple is a repeat player with a history of IPR claims being denied. This past injury was used to show the eminency of future injury resulting from the denial of the benefits of IPRs linked to the concrete interest possessed by an infringement defendant. Redressability and causation were met because there is a genuine possibility that the instructions would be changed in a way favorable to Apple in notice-and-comment rulemaking.  

The Federal Circuit may have reached a bit to find standing in an effort to effectively resolve concerns about a heavily used procedure: the IPR process. On remand, the district court might rightly decide that a traditional notice-comment rulemaking procedure is required to redress harms or prophylactically provide clarifications for the patent system that can accomplish the goals of using agency resources effectively. Allowing the frequent fliers of the IPR system to at least have an appearance of input in the procedure would create a process with more certainty and produce more long-term economic efficiency.  

13 thoughts on “IPRs and the APA: Review of Director’s Discretion to Initiate IPRs

  1. 3

    Apple v. Vidal, (Fed. Cir. Mar. 13, 2023):
    Either way [Director fiat or formal rulemaking], the current Sup. Ct. majority is said by Sup. Ct. experts to be much more open to challenges of federal agencies promulgating regulations which they think the enabling legislation did not Expressly authorize. An APA suit like this one is one way in which that issue can be raised, if the plaintiffs have sufficient “standing.” [Which was only narrowly granted to one of the parties in this case.]

      1. 3.1.1

        I would have posted my ideas about long Covid here, but knowing what happened to my hat with shield, I posted it on Facebook. That way it’s already out there without someone else, (if it turns out to be correct) claiming it for themselves.


          Facebook sounds like the perfect forum for you to express your opinions on long-covid.

  2. 2

    Two typos. First is in the OP, 1st par., “note-and-comment” should be “notice-and-comment”.

    Second is in the opinion, I think, but not sure, 35 U.S.C. 116 should be section 316. (That also carried through to the OP.)

  3. 1

    Yet inventors were denied standing in our parallel case, where the judge held that patent owners are not harmed by institution of an IPR. Thus the U.S. justice system holds that a multi-trillion dollar corporation may seek relief when they are limited to less than 3 vehicles to invalidate a patent which poses a liability of a fraction of a percent of their profit; while the court is closed to inventors who are compelled to a revocation proceeding that more often than not results in bankruptcy. This is corruption, not justice.

    1. 1.1

      1) I don’t see how the Fintiv standard harmed patent owners at all, so it’s not surprising that the Fifth Circuit–not the Federal Circuit–found no standing. My read of that case is that the plaintiffs argued that if the Fintiv opinion were thrown out for not going through notice-and-comment, then the PTO would have to promulgate rules through notice-and-comment (they would not have to), and those rules would be more stringent than Fintiv (not sure why), which would help patent owners avoid IPR institution. Even if the plaintiffs were involved in IPRs just as often as Apple and the other defendants, throwing out Fintiv is much less likely to redress any injury of plaintiffs because the PTO director doesn’t *have to* exercise her discretion to deny institution ever.
      2) Apple is constantly involved in probably dozens of IPRs, and I’d bet they make at least a few arguments on Fintiv/discretionary denial a year, if not more.
      3) Since the US Inventor plaintiffs and Apple both want the PTO to go through notice-and-comment for discretionary denial, aren’t their interests aligned here – and thus you should be happy?

      1. 1.1.1

        Your 1) seems entirely speculative and does not address the dichotomy present here (it does not depend on which side of the “v” one is on, now does it?)

        Your 2) while perhaps a legitimate statement in and of itself, does not establish standing in this case any more than that which was denied to that other side of the “v.”

        Your 3) quite misses the point, as while a certain operation may be in common, the arguments and rationales will most likely be vastly different. Also, this is far too glib as to the driving difference in treatment by the Court being its own issue.


          1) Yes it does depend what side of the “v.” the party is on because the point of Fintiv is to avoid institution even where the statutory merits test is satisfied. If Fintiv were thrown out, at least until a proper rule is put into effect, the PTO would have to make institution decisions solely based on the statutory test, rather than on discretionary grounds (as was the case pre-Fintiv). Plaintiffs instead argue that if Fintiv were junked, notice-and-comment might yield an even more restrictive rule. The latter ground is much more speculative – does anyone honestly think that would actually happen?

          2) Standing obviously takes account of whether a party is constantly battling the rule in question and therefore has a high likelihood of encountering it again. There’s no requirement that Apple could only bring this suit on direct appeal of a Fintiv denial. Procedurally, I’m not even sure they could. It would be some kind of bizarre APA suit in the form of a mandamus petition to CAFC after institution denial.

          3) Who cares about the arguments and rationales? Either way the effect is to throw out Fintiv while the PTO goes through notice-and-comment.


            1) no, no it does not so depend, given as the focus is procedural rather than content based.

            2) No, no it does not properly does so, as standing is required in a present action. That is rather the point. Further – and also directly on point to the disparity at issue, is that ANY such “other case” is exactly what the Court already said was not enough. The Court here cannot (legitimately) have it both ways.

            3) EVERY party cares. What may happen later can very much be affected by exactly whom has the ability to make their arguments.

      2. 1.1.2

        Lack of rules on discretionary denial jeopardizes our patent rights. If there were rules, we would know with more certainty whether our patents are likely to be eligible for AUA trials. It’s not just that Fintiv didn’t pass through NPRM, it’s that no procures have gone though NPRM, and that AIA and APA require rulemaking. Yet Apple and Google et al, on the other hand, don’t want rules.

    2. 1.3

      I agree with you, the granting of standing does not appear to be justice to me, and this is not good for patent holders in general. The frequent users of the system are those most often accused of infringing patents, and thus this sort of logic creates is likely to create a further bias toward invalidating patents. A patent system with unreliable patents does not seem like a good thing to me.

Comments are closed.