A company can spend years of research and tens of millions of dollars developing an invention, navigate the long road of patent prosecution, and emerge with a granted patent that looks bulletproof on paper. Then, in litigation, a single overlooked email, an undisclosed prior-art reference, or an inconvenient line in a regulatory filing surfaces -- and a court declares the entire patent unenforceable. Not one claim. Not just the tainted claim. The whole patent, often along with related patents in the same family. That is the doctrine of inequitable conduct, and the Federal Circuit itself famously called it the "atomic bomb" of patent law because of its all-or-nothing destructive power. Aventis Pharma S.A. v. Amphastar Pharmaceuticals, Inc., 525 F.3d 1334, 1349 (Fed. Cir. 2008) (Rader, J., dissenting).

This article explains, in plain language and in depth, what inequitable conduct is, where it comes from, how the modern legal standard works, and -- most importantly -- how inventors and the lawyers who represent them can avoid ever detonating it. It is written so that an inventor with no legal training, an in-house business lawyer, and a seasoned patent litigator can all follow it. Terms of art are defined the first time they appear, and worked hypotheticals (using invented companies such as "Acme Corp." and "Beacon Diagnostics") illustrate how the rules play out in practice. The hypotheticals are clearly labeled and are not based on real companies or events.

What You Will Learn

This article covers, in order: the duty of candor and good faith that every patent applicant owes the U.S. Patent and Trademark Office (USPTO) under 37 C.F.R. § 1.56; the history of the inequitable-conduct doctrine and how it grew into a litigation epidemic; the landmark en banc decision in Therasense, Inc. v. Becton, Dickinson & Co., 649 F.3d 1276 (Fed. Cir. 2011), which rewrote the standard; the two pillars of the modern test -- but-for materiality and specific intent to deceive -- and the demise of the old "sliding scale"; the narrow egregious misconduct exception that survives Therasense; the role of Kingsdown Medical Consultants, Ltd. v. Hollister Inc., 863 F.2d 867 (Fed. Cir. 1988) (en banc), in setting the intent floor; the aggressive use of litigation misconduct in Regeneron Pharmaceuticals, Inc. v. Merus N.V., 864 F.3d 1343 (Fed. Cir. 2017); the danger of infectious unenforceability spreading across a patent family; the Information Disclosure Statement (IDS) as the practical heart of compliance; the curative mechanism of supplemental examination under 35 U.S.C. § 257; the heightened pleading gate of Federal Rule of Civil Procedure 9(b) under Exergen; and a concrete, step-by-step compliance program. A FAQ and key-takeaways section round out the discussion.

The Foundation: The Duty of Candor and Good Faith

Patent prosecution is not an adversarial proceeding in the way a lawsuit is. When you apply for a patent, there is no opposing party in the room with the examiner challenging your invention. The examiner relies heavily on what the applicant tells them and on the references the applicant chooses to bring forward. Because of that lopsided, largely ex parte posture (a Latin term meaning a proceeding conducted for the benefit of one side, without the other present), the law imposes an affirmative obligation of honesty on everyone involved in the application. That obligation is the duty of candor and good faith.

The duty is codified in the USPTO's regulations at 37 C.F.R. § 1.56, often referred to simply as "Rule 56." It provides that each individual associated with the filing and prosecution of a patent application has a duty to disclose to the Office all information known to that individual to be material to patentability. The Manual of Patent Examining Procedure (MPEP), the USPTO's internal handbook, elaborates on the duty at MPEP §§ 2000.01 and 2001.04. The companion rule of professional conduct, 37 C.F.R. § 11.303, imposes parallel candor obligations on registered patent practitioners as a matter of attorney ethics, and a violation can be the basis for USPTO discipline (37 C.F.R. §§ 11.19(b)(1)(iv), 11.303(e), 11.804(c)). The same duty extends into ex parte reexamination under 37 C.F.R. § 1.555. (The separate, somewhat different candor regime governing proceedings before the Patent Trial and Appeal Board under 37 C.F.R. § 42.11 is beyond our scope here.)

The rationale is straightforward and worth stating plainly: a patent is a bargain with the public. In exchange for fully and honestly disclosing an invention, the inventor receives a time-limited right to exclude others from making, using, or selling it. (For the contours of that right, see our utility patent basics and patent basics -- a plain-English guide.) The USPTO grants that exclusive right on behalf of the public, and it can only do its job of separating genuinely new inventions from old or obvious ones if applicants play straight. When an applicant hides the ball, the entire bargain is corrupted -- which is why the remedy is so severe.

Who Owes the Duty

The duty of candor is not limited to the patent attorney signing the papers. Under 37 C.F.R. § 1.56(c), it applies individually to:

  • Each named inventor. Every inventor must also affirmatively acknowledge the duty in the inventor's oath or declaration filed with the application (37 C.F.R. § 1.63(c)).
  • Each attorney or patent agent who prepares or prosecutes the application.
  • Every other person who is substantively involved in the preparation or prosecution of the application and who is associated with the inventor, the assignee, or the attorney or agent. This can reach in-house counsel, scientific staff, regulatory personnel, licensing managers, and corporate officers who participate meaningfully in the process.

Critically, the duty runs throughout the entire pendency of the application, including after a notice of allowance issues and right up until the patent grants (37 C.F.R. § 1.56(a); MPEP § 2001.04). A person can satisfy the duty by disclosing material information to the prosecuting attorney or agent, who then takes on responsibility for getting it to the examiner (37 C.F.R. § 1.56(d)). In other words, a research scientist at Beacon Diagnostics does not have to file papers at the USPTO personally; she discharges her duty by telling the prosecuting attorney about the troublesome study she ran. But she must tell someone in a position to act -- silence is not an option once she knows the information is material.

What "Material to Patentability" Means

Not every scrap of information must be disclosed. The duty attaches only to information that is material -- a legal term meaning, in essence, important enough to affect whether a patent should issue. Rule 56(b) defines material information as information that is not cumulative (not duplicative of something already in the record) and that either:

  1. establishes, by itself or in combination with other information, a prima facie case that a claim is unpatentable (a prima facie case is one strong enough to require rebuttal if not contradicted); or
  2. refutes or is inconsistent with a position the applicant takes in arguing patentability or opposing a rejection.

Material information is by no means limited to classic prior-art patents and printed publications. The categories the USPTO and Federal Circuit recognize include:

  • Prior art under 35 U.S.C. § 102 -- earlier patents, published applications, products, publications, public uses, and sales that bear on whether the invention is new. (For how prior art shapes patentability arguments, see our guides to overcoming obviousness rejections and what constitutes patent infringement.)
  • Potential double-patenting references -- the applicant's own earlier or co-pending claims.
  • Enablement and written-description information under 35 U.S.C. § 112, such as data showing the invention does not work as broadly as claimed.
  • Inventorship and ownership information. Misrepresenting who actually invented something can itself be inequitable conduct (PerSeptive Biosystems, Inc. v. Pharmacia Biotech, Inc., 225 F.3d 1315, 1322 (Fed. Cir. 2000)).
  • Information from related applications and proceedings, including office actions, search reports, and rejections in the applicant's own foreign counterpart applications (MPEP § 2001.06(a)-(b)), and material from co-pending litigation (MPEP § 2001.06(c)).
  • Information from other government agencies, such as data submitted to the U.S. Food and Drug Administration that contradicts representations made to the patent examiner (see 87 Fed. Reg. 45,764 (July 29, 2022)).
  • Test results and data relating to the claimed invention, including unfavorable or failed experiments.
  • Statements and admissions, including any representation in the specification or arguments that is inconsistent with information already in the record, and any false or misleading statement requiring correction (MPEP §§ 2000.01, 2011).

One important practical point about scope: there is generally no affirmative duty to search for prior art (Rule 56 does not require a search). You only must disclose what you actually know to be material. But you cannot deliberately bury your head in the sand, and a search is almost always advisable, both to strengthen the patent and to identify what must be disclosed. The flip side, codified at 37 C.F.R. § 1.97(h), is equally important: disclosing a reference does not automatically make it material, and submitting information to the examiner is not an admission of materiality. That rule exists precisely to encourage over-disclosure without penalizing applicants for it. As a strategic matter, information that the examiner actually considers is also far less likely to support a successful later patent challenge, so generous disclosure both insures against inequitable conduct and hardens the patent against invalidity attacks.

A Brief History: How the Doctrine Became an Epidemic

Inequitable conduct did not begin as a regulatory rule. It grew out of the Supreme Court's "unclean hands" jurisprudence -- the equitable principle that a party who has acted dishonestly may not ask a court for relief. (For the broader family of equity-based patent defenses, see our overview of understanding equitable defenses -- laches, acquiescence, waiver, and equitable estoppel.) Three Supreme Court decisions from the 1930s and 1940s laid the groundwork: Keystone Driller Co. v. General Excavator Co., 290 U.S. 240 (1933); Hazel-Atlas Glass Co. v. Hartford-Empire Co., 322 U.S. 238 (1944); and Precision Instrument Manufacturing Co. v. Automotive Maintenance Machinery Co., 324 U.S. 806 (1945). Each involved serious fraud -- manufactured evidence, suppressed publications, a fabricated and falsely sworn article. In each the Court refused to enforce the patent. These were the doctrine's "egregious misconduct" ancestors, and they explain why the modern exception for affirmative fraud still exists.

Over the following decades, the Federal Circuit and its predecessor courts broadened the concept well beyond outright fraud. By the 1980s and 1990s, a defendant in nearly every patent infringement suit could be expected to plead inequitable conduct, often on thin evidence that the applicant had failed to cite some reference. The defense became, in the Federal Circuit's own words, "an absolute plague." Burlington Industries, Inc. v. Dayco Corp., 849 F.2d 1418, 1422 (Fed. Cir. 1988). Because it threatened the entire patent, drew the inventor's personal integrity into question, and could expose the case to additional consequences such as attorney's-fee awards and antitrust liability, the mere allegation carried enormous in terrorem settlement pressure even when the facts were weak. It also spread to the most routine cases: nearly every accused infringer pleaded it, and discovery ballooned as defendants combed prosecution files for any uncited reference.

The court tried to rein in the abuse with Kingsdown Medical Consultants, Ltd. v. Hollister Inc., 863 F.2d 867 (Fed. Cir. 1988) (en banc), which held that gross negligence alone does not establish intent to deceive. The conduct, viewed in light of all the evidence, must indicate sufficient culpability to require a finding of intent to deceive. Kingsdown set an important floor on the intent element, but the lower courts continued to apply a permissive "sliding scale" that let a strong showing of materiality substitute for a weak showing of intent, and the epidemic persisted. By 2011, the Federal Circuit was ready to operate.

The Therasense Earthquake

The watershed moment came in Therasense, Inc. v. Becton, Dickinson & Co., 649 F.3d 1276 (Fed. Cir. 2011) (en banc). Sitting en banc (meaning the full court, not a three-judge panel, decided the case, which signals the importance of the ruling), the Federal Circuit deliberately and substantially tightened the inequitable-conduct standard to "redirect a doctrine that has been overused to the detriment of the public." Id. at 1290. Therasense remains the controlling framework today, and every modern inequitable-conduct dispute is litigated against its requirements.

The Underlying Facts

The dispute is worth understanding because the facts illustrate the precise kind of inconsistency the doctrine targets. Therasense (then Abbott) had a patent on a blood-glucose test strip. During prosecution, to overcome a rejection based on one of its own earlier patents, Abbott argued to the U.S. examiner that a particular phrase in the earlier patent ("optionally, but preferably") meant the prior-art membrane was truly optional. Meanwhile, years earlier, in prosecuting the European counterpart to that very same earlier patent, Abbott's representatives had told the European Patent Office essentially the opposite -- that the membrane was required. Abbott never disclosed those contrary European statements to the U.S. examiner. The district court found the patent unenforceable, and the case went up to the full Federal Circuit. The lesson embedded in the facts is timeless: positions taken before foreign offices, and before any other tribunal, can come back to haunt you if they contradict what you tell the USPTO.

The New Standard: Two Independent Pillars

Therasense held that to prove inequitable conduct, the challenger must establish two separate elements, each by clear and convincing evidence (a demanding burden of proof higher than the ordinary "preponderance" standard used in most civil cases, requiring the fact-finder to have a firm conviction that the allegation is true):

Pillar One -- But-For Materiality. The withheld or misrepresented information must be but-for material, meaning the USPTO would not have allowed the claim had it been aware of the undisclosed reference. Id. at 1291. The court instructed trial judges to step into the examiner's shoes: applying the lower "preponderance of the evidence" standard the USPTO uses, and giving claims their broadest reasonable interpretation, would the examiner have allowed the claim if the truth had been disclosed? If yes, the information was not but-for material, and the inequitable-conduct defense fails on materiality alone. This was a deliberate rejection of the USPTO's older, far broader "reasonable examiner" materiality standard, under which information was material if a reasonable examiner would merely have considered it important. Notably, the USPTO has itself proposed harmonizing the Rule 56 materiality standard with Therasense's but-for test (see 81 Fed. Reg. 74,987 (Oct. 28, 2016); 76 Fed. Reg. 43,631 (July 21, 2011)), reflecting a broad consensus that materiality should be tied to actual effect on patentability.

Pillar Two -- Specific Intent to Deceive. The challenger must separately prove that the applicant acted with specific intent to deceive the USPTO. Building directly on Kingsdown, the court held that the accused infringer must show the patentee (1) knew of the reference, (2) knew it was material, and (3) made a deliberate decision to withhold it. Id. at 1290. Negligence -- even gross negligence -- is not enough. And in a passage that has decided many cases, the court held that when intent is proven circumstantially (as it almost always is, because few applicants confess), the inference of deceptive intent must be "the single most reasonable inference able to be drawn from the evidence." Id. (quoting Star Scientific, Inc. v. R.J. Reynolds Tobacco Co., 537 F.3d 1357, 1366 (Fed. Cir. 2008)). If there is an equally plausible innocent explanation for the omission, intent has not been proven. The patentee, importantly, has no obligation to offer a good-faith explanation until the challenger first carries its threshold burden on intent.

The Death of the Sliding Scale

Before Therasense, courts used a "sliding scale": a particularly strong showing of materiality could lower the quantum of intent the challenger had to prove, and vice versa. The two elements bled into each other. Therasense abolished this. The court held that materiality and intent are separate requirements, each of which must be independently established: "A district court should not use a 'sliding scale,' where a weak showing of intent may be found sufficient based on a strong showing of materiality, and vice versa." Id. at 1290. This single change is the most consequential practical effect of the decision. A defendant can no longer parlay a damning-looking omission into an inference of bad intent, or vice versa. The court must find both, separately, on clear and convincing evidence. In practice, this means a strong materiality record -- even a slam-dunk anticipatory reference -- buys the challenger nothing on intent, and a strong intent record buys nothing on materiality.

The Final Balancing of the Equities

Even after both pillars are established, inequitable conduct remains an equitable doctrine, so the court must weigh the equities to decide whether the patentee's conduct warrants the harsh sanction of unenforceability. In practice, once but-for materiality and specific intent are both proven, the equities will usually favor unenforceability, but the balancing step remains a formal part of the analysis. The Federal Circuit restated the full modern test -- (1) but-for materiality, (2) knowledge of the reference, (3) knowledge of its materiality, (4) a deliberate decision to withhold as the single most reasonable inference, and (5) a balancing of the equities -- in GS CleanTech Corp. v. Adkins Energy LLC, 951 F.3d 1310, 1324 (Fed. Cir. 2020).

A Worked Example (Hypothetical)

Consider Acme Corp., which files a patent application on a new battery cathode. During prosecution Acme's attorney argues that no prior reference teaches a particular cobalt-to-nickel ratio. Unbeknownst to the examiner, Acme's lead engineer is aware of a 2018 Japanese patent that discloses exactly that ratio, and Acme has a partial English translation in its files. Acme does not disclose it.

Under Therasense, the accused infringer must prove two things. First, but-for materiality: would the examiner have refused the claim had the Japanese reference been disclosed? If the reference truly discloses the same ratio in a way that anticipates or renders obvious Acme's claim, the answer is likely yes -- the reference is but-for material. Second, specific intent: was withholding the reference a deliberate decision to deceive, and is that the single most reasonable inference? Here the engineer's actual knowledge of the reference, the existence of a translation in Acme's files, and the attorney's affirmative argument that no such reference existed all point toward a deliberate withholding. If Acme cannot offer a credible innocent explanation -- for example, that the engineer reasonably believed the ratio in the reference was different, or that the reference was already cumulative to art of record -- a court could find inequitable conduct and render the entire patent unenforceable. Contrast this with a situation where Acme merely failed to find the reference in a search, with no one at the company actually aware of it: there, intent collapses, because you cannot deliberately withhold what you do not know you have. The hinge in nearly every real case is not whether the reference looks damaging, but whether the record proves that a specific human being knew it was material and chose to hide it.

The Egregious-Misconduct Exception

Therasense did not entirely eliminate liability for conduct that is not strictly but-for material. The court carved out a narrow exception for affirmative egregious misconduct. When the patentee engages in affirmative acts of egregious misconduct -- the paradigm being the filing of an unmistakably false affidavit or declaration -- the misconduct is considered material per se, and the challenger need not separately prove but-for materiality. Id. at 1292-93. The court reasoned that such affirmative fraud could not be tolerated regardless of its but-for effect, drawing the line back to the Supreme Court's Hazel-Atlas and Precision Instrument fraud cases.

The exception is genuinely narrow. The court emphasized that "neither mere nondisclosure of prior art references to the PTO nor failure to mention prior art references in an affidavit constitutes affirmative egregious misconduct." Id. at 1293. Simply not citing a reference will never qualify; the exception is reserved for affirmative, deliberate fabrication. The regulation itself preserves the same principle: 37 C.F.R. § 1.56(a) provides that no patent will be granted on an application "in connection with which fraud on the Office was practiced or attempted or the duty of disclosure was violated through bad faith or intentional misconduct." See also Transweb, LLC v. 3M Innovative Properties Co., 812 F.3d 1295, 1304 (Fed. Cir. 2016).

Two post-Therasense cases illustrate the exception in action:

  • In Intellect Wireless, Inc. v. HTC Corp., 732 F.3d 1339 (Fed. Cir. 2013), the inventor submitted a declaration falsely representing that he had actually built and demonstrated a working prototype of a wireless picture-phone, when no such demonstration ever occurred. A later "revised" declaration did not cure the falsehood because it did not expressly and clearly correct the prior misrepresentation. The Federal Circuit affirmed unenforceability under the egregious-misconduct exception. The case stands for an important corollary: to cure a false statement, the applicant must expressly call attention to both the error and the correction; a quiet, ambiguous walk-back is not enough. Id. at 1345; MPEP § 2011.
  • In Apotex Inc. v. UCB, Inc., 763 F.3d 1354 (Fed. Cir. 2014), the inventor (who was also the prosecuting attorney) made affirmative misrepresentations about a competitor's commercial product and submitted misleading experimental data to manufacture an inventive distinction. The court affirmed unenforceability, underscoring that affirmative misrepresentations -- not just omissions -- remain squarely actionable after Therasense.

The practical takeaway is stark: an inventor or attorney can survive an honest failure to cite a reference far more easily than an affirmatively false statement in a sworn declaration. Declarations are where applicants most often blow themselves up, because they are sworn, they are affirmative, and a false one is material per se.

Litigation Misconduct as Evidence of Intent: Regeneron v. Merus

One of the most striking developments since Therasense is the use of litigation misconduct -- bad behavior during the lawsuit itself, years after prosecution -- as circumstantial evidence of the deceptive intent that existed during prosecution. The leading case is Regeneron Pharmaceuticals, Inc. v. Merus N.V., 864 F.3d 1343 (Fed. Cir. 2017), reh'g denied, 878 F.3d 1041 (Fed. Cir. 2017), cert. denied, 139 S. Ct. 122 (2018).

In Regeneron, the district court found that during prosecution the patentee had withheld four prior-art references that were but-for material to the patent (which related to genetically modified mice for producing human antibodies). On the question of intent, however, the district court took an unusual route. Regeneron had engaged in extensive discovery misconduct during the litigation -- including improperly shielding documents and witnesses and violating the court's orders. As a sanction for that litigation misconduct, and because Regeneron's behavior obstructed the court's ability to assess what its prosecution-era intent had been, the district court drew an adverse inference of specific intent to deceive the USPTO. Id. at 1351, 1364. The Federal Circuit affirmed.

The lesson is sobering and counterintuitive: a patentee can lose its patent to inequitable conduct in part because of how its lawyers behaved in the courtroom years after the patent issued. As USPTO guidance underscores, a court may infer intent to deceive the Office "based on litigation misconduct outside of the prosecuting attorney's control." This means the original prosecution can be impeccable, yet the patent can still fall if the litigation team stonewalls discovery. For practitioners, Regeneron is a powerful reminder that candor obligations and clean-hands principles do not end when the patent issues -- they extend into any future enforcement. (For a deeper treatment of how these cases are built and rebutted, and how to develop the discovery record, see our companion piece on finding evidence of inequitable conduct in patent prosecution, and our broader comprehensive guide to patent infringement litigation.)

Other Pathways to a Finding

Beyond classic prior-art omissions, several recurring fact patterns have produced inequitable-conduct findings. Understanding them helps applicants recognize danger before it arises.

Misleading data and prophetic examples. Patent specifications sometimes include "prophetic examples" -- experiments that were not actually run but are predicted to work, which the rules permit if written in the future or present tense rather than the past tense (MPEP §§ 608.01(p), 2004). Presenting a prophetic example as though it were an actual, completed experiment, or selectively withholding failed experiments in a way that distorts the data, can support an inequitable-conduct finding. See Purdue Pharma L.P. v. Endo Pharmaceuticals Inc., 438 F.3d 1123, 1131 (Fed. Cir. 2006), and Novo Nordisk Pharmaceuticals, Inc. v. Bio-Technology General Corp., 424 F.3d 1347, 1362 (Fed. Cir. 2005). The tense of a single verb has decided cases: "the cathode was synthesized" implies a completed experiment, while "the cathode is synthesized" reads as predictive.

Inconsistent statements within the record and across agencies. In Belcher Pharmaceuticals, LLC v. Hospira, Inc., 11 F.4th 1345, 1354 (Fed. Cir. 2021), the patentee told the USPTO that a particular pH range was "critical" and unexpected while, in parallel FDA submissions, it had characterized the same range as old and well known. The chief science officer's knowledge of both contradictory positions supported a finding of inequitable conduct. Belcher is a vivid warning about the danger of telling one regulator one story and another regulator the opposite -- the same risk that animated the facts of Therasense in the foreign-prosecution context. For life-sciences companies in particular, FDA and USPTO filings must be reconciled before either is finalized.

Withholding related-application rejections. Failing to disclose adverse office actions, rejections, or search results from the applicant's own related or foreign-counterpart applications is a classic trap. If a sister application was rejected over a reference, that reference -- and often the rejection itself -- is exactly the kind of material information Rule 56 targets (MPEP § 2001.06(a)-(b)). This is especially dangerous when prosecuting a U.S. application while a parallel European or PCT application is being rejected over art the U.S. examiner never sees.

Inventorship misrepresentations. Deliberately naming or omitting an inventor to gain an advantage -- for example, to avoid an inventor's prior work being treated as prior art, or to manipulate ownership -- can be inequitable conduct (PerSeptive Biosystems, 225 F.3d at 1322). Getting inventorship right starts at the very beginning of the process; see our guidance on how to prepare an invention disclosure for your patent attorney. Note that innocent inventorship errors can be corrected without penalty under 35 U.S.C. § 256; it is the deceptive ones that are actionable.

Infectious Unenforceability: When One Bad Patent Spoils Many

Because inequitable conduct renders the entire patent unenforceable -- not merely the affected claim -- the consequences are already severe. But the doctrine can reach further still. Under the principle of infectious unenforceability (sometimes called "infectious inequitable conduct"), misconduct committed during the prosecution of one patent can render related patents unenforceable too, even if no misconduct occurred in the prosecution of those related patents.

The foundational case is Consolidated Aluminum Corp. v. Foseco International Ltd., 910 F.2d 804, 808-12 (Fed. Cir. 1990). The Federal Circuit held that inequitable conduct in one patent can spread to other patents in the same family where there is an "immediate and necessary relation" between the misconduct and the claims of the related patents. The phrase "immediate and necessary relation" comes from the unclean-hands tradition: the equitable taint extends only to claims that have a direct nexus to the original wrongdoing, not to every patent a company happens to own.

The practical mechanics matter. Patents are often filed as families -- a parent application spawns continuations, divisionals, and continuations-in-part, all sharing a common specification and priority chain. (For how those relationships are formed and managed, see our discussion of responding to patent office actions.) If the misconduct goes to subject matter, arguments, or data that carry through the family, the contamination can ride along the priority chain.

Worked example (hypothetical): Suppose Beacon Diagnostics commits inequitable conduct by submitting a false declaration in its parent application about clinical-trial results, and it then files three continuations that rely on those same results to support patentability of overlapping claims. A court could find not only the parent but all three continuations unenforceable, because the misconduct bears an immediate and necessary relation to the claims that issued from it. One false declaration can thus take down an entire portfolio segment, vaporizing years of prosecution investment in a single ruling.

The countervailing point for patentees: infectious unenforceability is not automatic. The challenger must establish the requisite nexus. Misconduct in a wholly unrelated application, or misconduct that does not touch the subject matter of the related claims, will not necessarily spread. But the doctrine is reason enough to treat candor compliance as a portfolio-wide discipline, not an application-by-application afterthought. The single most effective structural defense is to route material information identified in any one matter to every related matter automatically -- a point we return to in the compliance program below.

The Information Disclosure Statement: The Practical Heart of Compliance

In day-to-day practice, the duty of candor is discharged primarily through the Information Disclosure Statement (IDS) -- the formal vehicle by which an applicant submits known prior art and other material information to the examiner for consideration. Mastering IDS timing and content is the single most effective way to avoid inequitable-conduct exposure. The governing rules are 37 C.F.R. §§ 1.97 and 1.98.

The timing rules create a tiered structure, and the cost and procedural hurdles rise the later you file:

  • Early window (no fee, no statement). An IDS filed within three months of the application's filing date, or before the first office action on the merits (whichever is later), is considered as a matter of right -- no fee and no certification required (37 C.F.R. § 1.97(b); MPEP § 609.04(b)). This is the cleanest path, and best practice is to gather and submit known material art at the outset.
  • After the first office action (fee required). An IDS filed after the first office action on the merits but before a final action or notice of allowance requires either a fee or an applicant certification under § 1.97(e) (37 C.F.R. § 1.97(c)).
  • After final rejection or notice of allowance (fee and statement). An IDS filed at this late stage requires both a fee and a certification statement under § 1.97(d)-(e). The certification must state, for each item, either that it was first cited in a foreign counterpart application not more than three months earlier, or that no individual with a duty of candor knew of the information more than three months before the filing. The foreign-counterpart prong is not softened by any "lack of personal knowledge" qualifier (MPEP § 609.04(b)). If counsel cannot truthfully make the certification, counsel should not file the IDS; the better course is a Request for Continued Examination (RCE) with the IDS attached.
  • After issue-fee payment. The USPTO normally will not consider an IDS once the issue fee is paid. Options then include petitioning to withdraw the application from issue and filing an RCE with the IDS (37 C.F.R. §§ 1.114, 1.313(c)(2)), letting the patent issue and filing a continuing application with the IDS, or using the Quick Path IDS (QPIDS) pilot program, which lets an examiner consider a late IDS through a conditional RCE if the IDS does not warrant reopening prosecution. Supplemental examination (discussed below) is another route once the patent has issued.

The substantive content requirements under 37 C.F.R. § 1.98 are equally important. An IDS must list each item, provide a legible copy of each non-U.S.-patent reference (and, for foreign-language references, a concise explanation of relevance and any existing English translation), and identify the references on the required form so the examiner can consider and initial each one. A reference the examiner does not initial as "considered" has not been formally placed before the Office, which can undercut the candor benefit of submitting it -- so confirming the examiner's initials during the post-allowance file review is a real compliance step, not a clerical one (MPEP § 609.05).

Two recurring IDS pitfalls deserve emphasis:

Burying. Submitting a few critical references hidden within a haystack of hundreds of marginally relevant ones can itself be problematic. While there is no per se rule against large IDS submissions, deliberately obscuring the most important reference among a deluge of dross can support an inference of intent to deceive. The safer practice is to disclose generously but to flag genuinely critical references so the examiner is not misled by sheer volume.

Foreign-language references. If you have only a partial translation of a key foreign reference, submit the reference, submit the translation you have, and expressly flag its limitations. The Therasense facts themselves grew out of inconsistent positions taken before foreign offices, and Semiconductor Energy Laboratory Co. v. Samsung Electronics Co., 204 F.3d 1368 (Fed. Cir. 2000), found inequitable conduct where the applicant submitted a misleading partial translation of a Japanese reference that omitted the most damaging portions. Selectively translating only the favorable passages is precisely the kind of affirmative half-truth that draws an intent inference.

One useful infrastructure tool is the USPTO's Global Dossier, which aggregates the file histories of family-member applications across participating patent offices. Pulling foreign office actions and cited references through the Global Dossier is an efficient way to make sure that art surfacing in a foreign counterpart is timely routed into the U.S. IDS -- closing one of the most common candor gaps before it can ever become a litigation exhibit.

Finally, IDS practice intersects with patent term adjustment (PTA). Even a fully compliant IDS filed under § 1.97(b)-(e) can trigger a PTA reduction under 37 C.F.R. § 1.704(c)(6), (8), or (10), and the § 1.97(e) certification differs slightly from the § 1.704(d)(1) PTA "safe harbor" certification. Counsel should not let PTA optimization tempt them into delaying or omitting a disclosure -- a few days of patent term are never worth an unenforceability risk -- but should be aware that timely disclosure and term preservation are managed by distinct certifications.

Curing Problems: Supplemental Examination Under 35 U.S.C. § 257

What happens if, after a patent has already issued, the patentee discovers that something material was never disclosed during prosecution? Before 2011, the patentee's options were limited and risky. The America Invents Act created a powerful curative tool: supplemental examination under 35 U.S.C. § 257.

Supplemental examination allows a patent owner to request that the USPTO consider, reconsider, or correct information believed to be relevant to the patent. Crucially, unlike an ordinary ex parte reexamination, the information underlying a supplemental examination request is not limited to prior-art patents and printed publications -- it may be directed to any type of patentability question (35 U.S.C. § 257(b)), which is exactly what makes it suited to candor problems involving declarations, data, or inconsistent statements. The key benefit is statutory and substantial: under § 257(c)(1), a patent shall not be held unenforceable on the basis of conduct relating to information that was considered, reconsidered, or corrected during a supplemental examination proceeding. In plain terms, supplemental examination can immunize a patent against an inequitable-conduct charge premised on the information that was put before the Office through the proceeding. It is, in effect, a confessional that washes away the prosecution sin -- if done in time and done right.

The mechanics are as follows. The patent owner files a request identifying each item of information and explaining its relevance (up to twelve items per request). Within three months, the USPTO determines whether any item raises a substantial new question of patentability (an SNQ -- the same threshold used for reexamination, meaning a question a reasonable examiner would consider important, a lower bar than a prima facie case of unpatentability). If it does, the USPTO orders an ex parte reexamination to resolve the question; if not, the proceeding concludes and a certificate issues. Either way, the patent then enjoys the § 257(c) protection as to the information considered.

There are important limitations that counsel must respect:

  • Timing. The supplemental examination must be requested before the inequitable-conduct allegation is pleaded with particularity in a civil action or asserted in a notice under the Hatch-Waxman (Paragraph IV) framework, and any resulting reexamination must conclude before the action is brought (§ 257(c)(2)). You cannot wait until you are sued and then race to the USPTO -- the protection is forfeited if you are already in the crosshairs.
  • No protection for prior allegations. The immunizing effect does not apply to defenses already raised before the request, or to allegations in an already-filed action.
  • Fraud referral. If the USPTO becomes aware that material fraud was practiced, it may refer the matter to the U.S. Attorney General (§ 257(e)). Supplemental examination cleanses inequitable conduct, but it is not a get-out-of-jail-free card for criminal fraud or Walker Process antitrust exposure.

Supplemental examination sits alongside the patent owner's other post-grant correction tools -- patent owner-requested ex parte reexamination and reissue under 35 U.S.C. § 251. All three can address prior-art and patentability issues, but they differ in important ways. A reissue can broaden (within two years) or narrow claims and correct a variety of errors, but it does not carry § 257's express statutory immunity from inequitable-conduct attacks, and the reissue oath itself can become a new candor minefield. Ordinary ex parte reexamination is limited to questions arising from patents and printed publications and likewise lacks the anti-unenforceability shield. Supplemental examination is uniquely suited to candor problems precisely because of its breadth (any patentability question) and its § 257(c) immunity. Where a patentee discovers a candor gap in a valuable, in-force patent, supplemental examination is usually the first option to evaluate -- ideally well before any litigation is on the horizon, because the window slams shut the moment an inequitable-conduct allegation is pleaded.

The Litigation Gateway: Rule 9(b) Pleading Under Exergen

Even a meritorious-sounding inequitable-conduct theory must clear a procedural gate before it ever reaches discovery or trial. Because inequitable conduct sounds in fraud, an accused infringer must plead it with the particularity required by Federal Rule of Civil Procedure 9(b). The controlling case is Exergen Corp. v. Wal-Mart Stores, Inc., 575 F.3d 1312 (Fed. Cir. 2009).

Exergen requires the pleading to identify the specific who, what, when, where, and how of the alleged material omission or misrepresentation. Id. at 1327-29. Concretely, the defendant must name the particular individual who owed and breached the duty (not just "the applicant" or "Acme"), identify the specific reference or specific false statement, allege where the material information appears (for example, which column and lines of the withheld reference), and explain why it is material and not cumulative to the art already before the examiner. The pleading must further allege facts from which the court can reasonably infer that the named individual (1) knew of the withheld material information and (2) withheld or misrepresented it with a specific intent to deceive the USPTO. Conclusory allegations that the applicant "should have known" about a reference, or that the omission "must have been" intentional, will not survive a motion to dismiss.

This pleading discipline, layered on top of the demanding Therasense substantive standard, has done much of the work of shrinking the inequitable-conduct "plague." A defendant can no longer reflexively tack the defense onto every answer; it must come to court with named names and specific facts. For patent owners, Exergen is the first line of defense -- a well-aimed motion to dismiss or to strike an inadequately pleaded inequitable-conduct allegation can remove the cloud early, before it drives up the cost and risk of the case.

Strategic Considerations for Litigators

For litigators on either side of an inequitable-conduct dispute, Therasense, Exergen, and their progeny shape strategy from the pleadings forward.

Building intent circumstantially. Because direct evidence of intent is rare, the accused infringer's case is built from documents: lab notebooks, internal emails debating whether to cite a reference, foreign-prosecution files taking contrary positions, FDA submissions that conflict with USPTO arguments (the Belcher pattern), and -- per Regeneron -- the patentee's own litigation conduct. The patentee's best defenses are a clean prosecution record, contemporaneous documentation of good-faith judgment calls, and, where applicable, a credible innocent explanation that makes deceptive intent something less than the single most reasonable inference. Remember that under Therasense the patentee need not even offer that explanation until the challenger first carries its threshold burden.

The interplay with attorney's fees and antitrust. A finding of inequitable conduct can ripple outward. It can render a case "exceptional" and justify a fee award under 35 U.S.C. § 285, and -- where the patent was both procured by knowing and willful fraud and asserted with anticompetitive effect -- it can give rise to so-called Walker Process antitrust liability (Walker Process Equipment, Inc. v. Food Machinery & Chemical Corp., 382 U.S. 172 (1965)). Importantly, Walker Process fraud is a higher bar than inequitable conduct: it requires the rigorous knowing-and-willful-fraud showing, so not every inequitable-conduct finding will support an antitrust counterclaim. These collateral consequences are part of why the doctrine remains feared even after Therasense narrowed it.

Privilege management. Because intent turns on what the prosecution and litigation teams knew and decided, inequitable-conduct disputes invariably press against the attorney-client privilege and work-product protection. Asserting a good-faith-reliance defense can waive privilege over prosecution communications, so the decision to put counsel's advice in issue must be made deliberately and early.

A Practical Compliance Program

The best defense against inequitable conduct is a disciplined, repeatable compliance program that makes candor the default rather than an afterthought. The following synthesizes USPTO guidance and Federal Circuit case law into concrete steps for inventors, in-house counsel, and outside prosecution attorneys.

Step 1: Educate Everyone Who Touches the Application

Because the duty applies individually to inventors, attorneys, and substantively involved staff, the program begins with education. Outside counsel should give every named inventor and relevant employee a plain-language explanation of the duty of candor: what must be disclosed, who is covered, that the duty continues until grant, and that the consequence of a violation is loss of the entire patent (and possibly the family). The inventor's declaration's candor acknowledgment under 37 C.F.R. § 1.63(c) should be presented as a real obligation, not boilerplate. Best practice is to repeat a short candor reminder in every substantive written communication with the client during prosecution, so the obligation stays front of mind as new information surfaces.

Step 2: Gather and Track Material Information Systematically

Inventors should be instructed to tell their attorney about all potentially relevant information -- prior products, publications, public uses, offers for sale, testing (including failed experiments), competitor activity, and any related litigation or regulatory filings. Even seemingly unfavorable information must surface; the failed experiment you hide is precisely the data a court will use against you. The company should maintain a centralized docket -- backed by tools such as the USPTO's Global Dossier for foreign family members -- that tracks, across the whole patent family, every reference cited in any related application, every office action in foreign counterparts, every international search report, and every relevant regulatory submission, so that material information identified in one matter is automatically routed to all related matters. This systematic cross-pollination is the single best defense against both ordinary candor lapses and infectious unenforceability. (Good information hygiene here also dovetails with broader IP risk management; see conducting freedom-to-operate analysis for new products.)

Step 3: Draft Compliant Applications

When preparing the specification, counsel should avoid mischaracterizing the prior art, avoid presenting prophetic examples as completed experiments (keep them in the present or future tense), and avoid cherry-picking favorable data while suppressing unfavorable results. Declarations and affidavits deserve special scrutiny because they are the vehicle through which the egregious-misconduct exception most often bites: every factual assertion in a declaration must be literally true and must not create a false impression by omission. The duty of candor also works in tandem with the practitioner's duty of reasonable inquiry and the implied certification that accompanies every paper filed with the Office.

Step 4: File Timely, Complete IDSs

The program should ensure that an IDS disclosing all known material information is filed within the early, fee-free window, and that supplemental IDSs are filed promptly whenever new material information comes to light -- including after an office action, after a foreign counterpart is examined, or after relevant litigation produces new prior art. Counsel should verify that the examiner has initialed each reference as considered, and should disclose generously while flagging genuinely critical references rather than burying them. After a notice of allowance, run a deliberate candor review confirming that all known material art was disclosed and that no uncorrected misstatement remains in the file.

Step 5: Correct Errors Expressly

If a misstatement or misleading impression is discovered anywhere in the record -- the specification, an argument, or a declaration -- the program must call for an express correction that identifies both the error and the corrected information. Under Intellect Wireless, a quiet or ambiguous walk-back does not cure a prior misrepresentation; the correction must be unmistakable (MPEP § 2011).

Step 6: Document Good Faith

Throughout, the file should reflect diligence and good faith. When a close call is made about whether a reference is material or cumulative, document the reasoning. When the team debates whether to disclose something and decides to err on the side of disclosure, that record is gold in later litigation, because it directly rebuts the "single most reasonable inference" of deceptive intent that Therasense requires. Courts reward visible diligence; an internal memo weighing both sides and choosing disclosure is often worth more than any after-the-fact testimony.

Step 7: Have a Post-Grant Cure Plan

Finally, the program should include a protocol for what to do if a candor gap is discovered after grant: evaluate supplemental examination under § 257 promptly, while the patent is still clear of litigation and before any inequitable-conduct allegation has been pleaded, so the statutory immunity remains available. Waiting until a lawsuit is filed forfeits the most valuable cure the system offers.

Frequently Asked Questions

Is failing to cite a single prior-art reference enough to invalidate a patent? Not by itself. After Therasense, the challenger must prove both that the reference was but-for material (the claim would not have issued had it been disclosed) and that someone with a duty of candor knew of the reference, knew it was material, and deliberately decided to withhold it with intent to deceive -- all by clear and convincing evidence. A mere omission, without proof of that deliberate, deceptive state of mind, will not support a finding. The exception is affirmative egregious misconduct, such as a knowingly false declaration, which is treated as material per se. And as a procedural matter, the allegation must first be pleaded with particularity under Rule 9(b) and Exergen just to survive a motion to dismiss.

Does disclosing a reference admit that it is material or that my invention is unpatentable? No. Under 37 C.F.R. § 1.97(h), submitting information to the USPTO is not an admission that the information is material or that it is prior art. This rule is specifically designed to encourage applicants to disclose generously without fear that doing so will be used against them. When in genuine doubt, the safer course is to disclose -- it both discharges the duty and tends to strengthen the patent against later validity challenges.

What is the difference between inequitable conduct and invalidity? Invalidity means a patent claim should never have issued -- for example, because it was anticipated by prior art or was obvious. It is assessed claim-by-claim. Inequitable conduct is different: it concerns how the applicant behaved before the USPTO, and when proven it renders the entire patent unenforceable regardless of whether the claims are otherwise valid. A perfectly valid invention can yield an unenforceable patent if the applicant deceived the examiner.

Can inequitable conduct in one patent affect my other patents? Yes, potentially. Under the doctrine of infectious unenforceability (Consolidated Aluminum Corp. v. Foseco International Ltd.), misconduct in the prosecution of one patent can render related patents unenforceable where there is an "immediate and necessary relation" between the misconduct and the claims of the related patents. This is a strong reason to manage candor compliance across an entire patent family rather than application by application.

If I discover a candor problem after my patent issues, can I fix it? Often, yes -- through supplemental examination under 35 U.S.C. § 257. If you ask the USPTO to consider, reconsider, or correct the relevant information before any inequitable-conduct allegation is raised against you, the patent cannot later be held unenforceable on the basis of that information. The timing is critical: the window closes once an inequitable-conduct allegation has been pleaded with particularity (or asserted in a Hatch-Waxman notice), so the cure must be pursued proactively. Supplemental examination is broader than ordinary reexamination because it can address any patentability question, not just prior art.

Do inventors personally face consequences, or just the company? Both can be affected. The patent itself becomes unenforceable, which harms the owner. But individuals who participated in the misconduct can face additional consequences: registered patent practitioners can be subject to USPTO disciplinary proceedings under 37 C.F.R. Part 11, and in extreme cases of knowing and willful fraud, antitrust (Walker Process) or even criminal exposure can attach. The duty of candor is personal, and so are some of its consequences.

How is inequitable conduct decided -- by a jury or a judge? Inequitable conduct is an equitable defense, so it is tried to the judge, not the jury, although a jury may make advisory findings on underlying facts. The judge applies the clear-and-convincing standard, makes the materiality and intent findings, and ultimately balances the equities.

Key Takeaways

Inequitable conduct remains the most catastrophic defense in patent litigation precisely because it destroys the entire patent and can spread to a family. The doctrine rests on a simple premise -- that patent applicants must be honest with the USPTO -- codified in the duty of candor under 37 C.F.R. § 1.56. Since Therasense v. Becton, Dickinson in 2011, the bar to prove it has been demanding: a challenger must establish, separately and by clear and convincing evidence, both but-for materiality and specific intent to deceive, with the old forgiving "sliding scale" abolished and the inference of intent required to be the single most reasonable one. A narrow exception survives for affirmative egregious misconduct such as false declarations. Kingsdown set the intent floor; Regeneron v. Merus showed that even later litigation misconduct can supply the inference of prosecution-era intent; Consolidated Aluminum shows how the taint can infect a whole portfolio; and Exergen erects a Rule 9(b) pleading gate that screens out thin allegations at the threshold.

For inventors and companies, the practical message is reassuring but demanding: you are unlikely to lose a patent to an honest mistake, but you must build a real culture of candor. Disclose known material information generously and on time through complete IDSs; reconcile your USPTO arguments with your FDA and foreign-office positions; never mischaracterize data or submit a false declaration; document your good-faith judgment calls; manage candor across the whole patent family; and, if a problem surfaces after grant, move quickly to supplemental examination under 35 U.S.C. § 257 before any allegation is raised. Candor is not merely an ethical nicety -- it is the price of an enforceable patent.

Related Articles

Disclaimer: This article is provided by mclaw.io for general informational purposes only and does not constitute legal advice, nor does it create an attorney-client relationship. Inequitable conduct and the duty of candor involve fact-intensive judgments and evolving case law; the consequences of a misstep are severe. Readers should consult qualified patent counsel about the specific facts of their situation before taking or refraining from any action.