Patent reviews

IPR estoppel: the trade-off behind the cheapest invalidity tool

File your strongest art, file it well — or don't file at all if your strongest grounds aren't patents and publications. Estoppel is what makes IPR powerful for the system and dangerous for the defendant who underprepares.

5 min read · Updated Apr 29, 2026

Inter Partes Review at the USPTO PTAB is the cheapest viable path to kill a weak patent — $200,000–$500,000 against $3M-$5M in district court, on a 12-month timeline, under a lower preponderance burden of proof. (See Patent reviews for the full mechanics.)

The trade-off is estoppel. Once the PTAB issues a final written decision in your IPR, you are barred from raising in district court any prior-art ground you "raised or reasonably could have raised" during the IPR. File the wrong art at the wrong time and you can lose your district-court invalidity defenses without ever winning at the PTAB.

This piece is the operating-manual version of estoppel — what triggers it, what survives, how to file without burning your fallbacks.

What § 315(e) actually says

35 U.S.C. § 315(e)(2) is the operative statute. The petitioner in an IPR (or the real party in interest, or any privy) "may not assert ... in a civil action" any ground for invalidity "that the petitioner raised or reasonably could have raised during that inter partes review."

Three things follow:

  1. The estoppel attaches to the petitioner. Including affiliates and anyone in privity. Joint defense groups have to be careful — joining an IPR can rope you into the estoppel.
  2. It triggers on a final written decision — not on institution, not on settlement, not on denial of institution. If the PTAB never institutes, no estoppel. If you settle pre-FWD, no estoppel.
  3. The scope is "raised or reasonably could have raised." Not just the grounds you actually argued — also the grounds you could have argued with reasonable diligence but didn't.

What "reasonably could have raised" really means

This is where the trap is. The Federal Circuit has held that the phrase covers any patent-or-printed-publication prior art that a reasonably diligent searcher would have found before filing the petition. California Institute of Technology v. Broadcom Ltd. (Fed. Cir. 2022) is the modern touchstone — it broadened the doctrine and made it harder to slip new prior art past estoppel.

Practical consequence: if you file an IPR with three references and lose, you generally cannot later argue invalidity in district court using a fourth reference that was findable at the time. The defendant is stuck with infringement-only defenses (claim construction, non-infringement, damages limitations) — which often aren't enough.

What estoppel does NOT cover

Estoppel is real but not total. Several invalidity grounds survive an IPR loss:

  • § 101 (subject matter / Alice). IPR can't reach § 101 in the first place, so § 101 challenges in district court are unaffected.
  • § 112 (written description, enablement, indefiniteness). Outside IPR's prior-art scope.
  • Prior-use / on-sale bar. § 102(a) public-use and on-sale prior art that isn't itself a patent or printed publication is outside IPR's reach. Limited but real.
  • Inequitable conduct. A separate defense based on patentee misconduct during prosecution.
  • Best-mode (pre-AIA patents only). Available only for patents filed before March 16, 2013.

A careful petitioner picks IPR when their strongest invalidity story rests on patents and publications, and keeps non-IPR-able defenses in reserve for district court.

Strategic implications

Three rules for filing IPR without burning yourself:

1. File your strongest art

The petition is the bet. The PTAB rejects petitions that scatter weak combinations of references — and the estoppel still attaches to whatever you raised. File the two or three strongest references you have. Don't hold them back for district court.

2. Don't file when your real defense is § 101 or non-infringement

If your strongest theory is that the patent is an abstract idea, IPR can't help — § 101 isn't an IPR ground. File a Rule 12(c) motion in district court instead. Same for non-infringement — IPR doesn't decide infringement. If your case is "we don't infringe" rather than "the patent is invalid," IPR adds risk without upside.

3. Coordinate with co-defendants on joint petitions carefully

Multiple defendants in the same suit sometimes file a single IPR jointly to share costs. Estoppel attaches to all the joint petitioners. If one defendant has a stronger non-infringement story, joining the IPR may foreclose other defendants' cleaner exits.

Workarounds

When estoppel risk is too high to justify the IPR upside, a few alternatives:

  • Ex parte reexamination (covered in Patent reviews). No estoppel attaches to the requester — they drop out after filing. Outcomes are noticeably more pro-patentee than IPR, but for clear prior-art cases the reexam can still kill the patent without burning IPR fallback grounds.
  • District-court summary judgment on § 101 or § 112. Stay out of the PTAB entirely and target the patent on grounds IPR can't reach.
  • Anonymous third-party reexamination. Coordinate with another party to file the reexam so your name isn't on it; estoppel attaches to that party, not you.

These workarounds have downsides (cost, slower timelines, narrower reach), but they preserve the IPR as a future option.

What's changed recently

The Federal Circuit has been steadily expanding IPR estoppel since the Caltech decision. Subsequent panels have applied estoppel to non-petition grounds the petitioner could have included, and to grounds raised by privy parties in different proceedings. The trend is toward a stricter "you had your shot" rule.

Defendants who file IPR in the post-Caltech era should assume the broadest possible estoppel applies — and design their petitions accordingly. The cost-effective IPR is the strong, well-prepared IPR. Half-petitions invite the worst outcome: lose at PTAB, also lose your fallback in court.

Bottom line

IPR estoppel is the price of cheap invalidity. The system trades district-court flexibility for PTAB efficiency — and the trade is favorable when your invalidity case is strong and entirely about patents-and-printed-publications, unfavorable when your strongest defense is § 101, non-infringement, or rests on non-patent prior art. File deliberately.

Related: Patent reviews: IPR, PGR, reexam, Prior art and § 102, Obviousness under § 103.

This article is for general education and is not legal advice.