PTAB (IPR / PGR)

Inter partes review is an administrative trial in which the Patent Trial and Appeal Board reconsiders whether an already-issued patent should have been granted. A petitioner, typically someone accused of infringement, asks three administrative patent judges to cancel specific claims as anticipated or obvious. The Board, not a jury, decides.

Congress created it in the Leahy-Smith America Invents Act of 2011, and it went live on September 16, 2012. The design goal was blunt: give the public a way to kill bad claims without spending years and millions in federal court. It worked well enough that IPR reshaped the economics of patent enforcement in the United States, and it has been fought over politically ever since.

What the PTAB can and cannot consider in an IPR

The single most important limitation sits in 35 U.S.C. § 311(b). An IPR petition may raise only novelty under § 102 and obviousness under § 103, and only on the basis of patents and printed publications. Nothing else is on the table.

That excludes a great deal. You cannot argue in an IPR that a claim is an ineligible abstract idea under § 101. You cannot argue indefiniteness, lack of enablement, or inadequate written description under § 112. You cannot rely on a prior public use, a prior sale, or a physical product that was never described in a document. Those defenses survive for district court, which is one reason IPR rarely ends a dispute by itself.

Post-grant review, the sibling proceeding under §§ 321 to 329, has no such fence. A PGR petition may raise almost any ground of invalidity, including eligibility and § 112 problems. The price of that breadth is a narrow window, discussed next.

The deadlines that decide which door you use

Three clocks govern access.

• The nine-month line. PGR must be filed within nine months of issuance (§ 321(c)). IPR may not be filed until nine months after issuance, or until any PGR terminates (§ 311(c)). They are sequential, not parallel.
• The one-year bar. Under § 315(b), an IPR petition is time-barred if filed more than one year after the petitioner, a real party in interest, or a privy was served with a complaint alleging infringement. This is the deadline that ends most would-be challenges. Miss it and the door closes permanently, regardless of how strong the prior art is.
• The first-inventor-to-file limit. PGR reaches only patents whose claims have an effective filing date on or after March 16, 2013. Older patents were never eligible.

A related transitional proceeding, covered business method review, was a transitional program that sunset on September 16, 2020 and no longer accepts petitions.

Why petitioners find the PTAB friendlier than a courtroom

Two structural differences do most of the work.

First, the burden of proof. In district court an issued patent carries a statutory presumption of validity, and the challenger must prove invalidity by clear and convincing evidence. At the PTAB, § 316(e) sets the burden at a preponderance of the evidence. The same prior art can lose in court and win before the Board without anyone being inconsistent.

Second, the decision-maker. The panel consists of administrative patent judges with technical degrees, not lay jurors. Whether that is a feature or a defect depends entirely on which side of the caption you sit on.

Claim construction used to be a third asymmetry. The Board originally applied the broadest reasonable interpretation, which made claims easier to read onto prior art. The Supreme Court blessed that choice in Cuozzo Speed Technologies v. Lee (2016), but the USPTO changed the rule by regulation effective November 13, 2018, and the Board now applies the same Phillips standard federal courts use. That alignment removed a long-standing complaint that a claim could mean two different things in two forums at once.

Institution: discretionary, all-or-nothing, and largely unreviewable

A petition does not become a trial automatically. The Director, acting through the Board, decides whether to institute. The threshold for IPR is a “reasonable likelihood that the petitioner would prevail” on at least one challenged claim (§ 314(a)); for PGR it is “more likely than not” (§ 324(a)).

Three features of institution surprise people:

• It is discretionary, not mandatory. Meeting the threshold earns you consideration, not a trial. The Board may decline for reasons unrelated to the merits.
• It is not appealable. Section 314(d) makes the institution decision “final and nonappealable,” and Cuozzo (2016) and Thryv, Inc. v. Click-to-Call Technologies (2020) read that bar broadly, the latter holding that even a disputed § 315(b) time-bar ruling escapes review.
• It is all-or-nothing on claims. In SAS Institute Inc. v. Iancu (2018), the Supreme Court held that if the Board institutes, it must address every claim the petitioner challenged. Partial institution is gone. The Board subsequently extended the same logic to every ground raised.

The discretionary piece is where the policy fight lives. Under Apple Inc. v. Fintiv, Inc., a 2020 precedential decision, the Board weighs factors such as the trial date in a parallel district court case, investment already made there, and overlap of issues, and may deny an otherwise meritorious petition simply because a court is likely to get there first. Fintiv practice has swung with successive USPTO administrations. A June 2022 interim memorandum narrowed discretionary denials, telling the Board not to deny on Fintiv grounds where the petitioner filed a broad Sotera stipulation or presented compelling evidence of unpatentability. The USPTO rescinded that memorandum on February 28, 2025, which restored the broader six-factor test and widened discretionary denials again. Anyone relying on a settled rule here should check the Board’s current guidance rather than a secondhand summary.

Estoppel: what you give up by filing

Section 315(e) is the trade. Once the Board issues a final written decision on a claim, the petitioner and its privies may not thereafter assert, in the USPTO, in a district court, or before the International Trade Commission, that the claim is invalid on any ground raised or that reasonably could have been raised during the IPR.

“Reasonably could have been raised” has been read to cover prior art a skilled searcher conducting a diligent search would have found, which is broader than what the petitioner actually located. The estoppel attaches per claim and only after a final written decision, so a denied institution or a settlement before decision does not trigger it.

The limit follows from § 311(b). Because IPR could never have reached prior public uses, on-sale bars, § 101 eligibility, or § 112 defects, estoppel cannot bar those arguments later. A defendant who loses an IPR on printed publications may still attack the same claim in court on a system prior art theory.

The constitutional and structural challenges, and how they came out

The system survived two serious attacks.

In Oil States Energy Services v. Greene’s Energy Group (2018), the Supreme Court held that a granted patent is a public franchise and that reconsidering it is a matter of public rights Congress may assign to an agency. IPR therefore violates neither Article III nor the Seventh Amendment. The Court expressly reserved due process and takings questions, and Justice Gorsuch dissented on the ground that a property right once granted should be revoked only by a court.

In United States v. Arthrex, Inc. (2021), the Court held that administrative patent judges were exercising authority reserved to principal officers without Senate confirmation, an Appointments Clause defect. The remedy was structural rather than fatal: the Director must be able to review and rescind Board decisions on his or her own initiative. That is why Director Review now exists as a step between the final written decision and appeal to the Federal Circuit.

Appeals from final written decisions go to the U.S. Court of Appeals for the Federal Circuit under § 319. The institution decision itself does not.

Frequently asked questions

What is inter partes review? Inter partes review is an administrative trial at the Patent Trial and Appeal Board in which a petitioner asks a panel of three judges to cancel claims of an issued patent as anticipated or obvious. Under 35 U.S.C. § 311(b) the challenge may rely only on patents and printed publications. It was created by the America Invents Act and became available on September 16, 2012 as a faster, cheaper alternative to litigating invalidity in district court.

What is the difference between IPR and PGR? Timing and grounds. Post-grant review must be filed within nine months of issuance and can raise almost any invalidity ground, including eligibility under § 101 and indefiniteness or lack of enablement under § 112. Inter partes review opens only after that nine-month window closes and is limited to novelty and obviousness based on patents and printed publications. PGR also reaches only patents examined under the first-inventor-to-file rules.

How long does an IPR take and what does it cost? The Board must issue a final written decision within one year of institution, extendable by six months for good cause, so the whole proceeding usually runs about eighteen months from petition to decision. USPTO fees alone run into the tens of thousands of dollars, and total cost through a final decision commonly reaches several hundred thousand. District court patent litigation frequently costs several million.

What is IPR estoppel? Under 35 U.S.C. § 315(e), once the Board issues a final written decision, the petitioner and its privies may not later argue in the USPTO, a district court, or the ITC that a claim is invalid on any ground that was raised or reasonably could have been raised in the petition. Because IPR is limited to patents and printed publications, estoppel does not block later defenses based on prior public use, on-sale activity, or § 101 eligibility.

Authorities and sources

• 35 U.S.C. § 311: Inter partes review (Cornell LII)
• 35 U.S.C. § 314: Institution of inter partes review (Cornell LII)
• 35 U.S.C. § 315: Relation to other proceedings and estoppel (Cornell LII)
• 35 U.S.C. § 316: Conduct of inter partes review (Cornell LII)
• 35 U.S.C. § 319: Appeal (Cornell LII)
• 35 U.S.C. § 321: Post-grant review (Cornell LII)
• 35 U.S.C. § 324: Institution of post-grant review (Cornell LII)
• USPTO: Patent Trial and Appeal Board
• USPTO fee schedule
• Cuozzo Speed Technologies, LLC v. Lee, 579 U.S. 261 (2016) (Cornell LII)
• SAS Institute Inc. v. Iancu, 584 U.S. 357 (2018) (Cornell LII)
• Oil States Energy Services, LLC v. Greene’s Energy Group, LLC, 584 U.S. 325 (2018) (Cornell LII)
• Thryv, Inc. v. Click-to-Call Technologies, LP, 590 U.S. 45 (2020) (Cornell LII)
• United States v. Arthrex, Inc., 594 U.S. 1 (2021) (Cornell LII)

Going further: Should you challenge a patent at the PTAB? .

This page is general legal information, not legal advice, and it does not create an attorney-client relationship.