The scope of IPR estoppel – system prior art, obviousness combinations, and separate grounds under 35 U.S.C. § 315(e)

Inter partes review (IPR) is an expedited forum for adjudicating patent invalidity challenges. Counterbalancing this administrative proceeding is an estoppel provision, 35 U.S.C. 315(e) that prohibits IPR petitioners who lose an IPR from re-litigating whether the patent “is invalid on any ground that the petitioner raised or reasonably could have raised during that inter partes review.” Section 315 thus bars some (but not all) subsequent invalidity challenges.

By design, IPRs only allow petitioners to raise a limited type of invalidity challenge. IPRs are restricted to invalidity challenges based on “a ground that could be raised under section 102 or 103,” ie, obviousness or anticipation. But IPRs do not allow all obviousness and anticipation grounds: section 102 and 103 challenges may be made “only on the basis of prior art consisting of patents or printed publications.” Challenges based on prior art showing “the claimed invention…in public use, on sale, or otherwise available to the public” may not form the basis of an IPR.

Some applications of the IPR estoppel statute are uncontroversial. Courts readily bar grounds that were actually raised and litigated in an IPR. The Federal Circuit has also barred grounds based on prior art references that were not litigated if the IPR petitioners “were aware of the prior art references” before filing the IPR and voluntarily elected not to include those references in the IPR petition.[1]

Courts have disagreed on two core issues regarding the scope of the IPR estoppel statute: (1) how the estoppel provision applies to newly discovered patents and publications and (2) how it applies to prior art that is not eligible for inclusion in an IPR petition but which is substantively similar to art that is eligible (eg, a product [ineligible as an IPR ground] and the product’s printed manual [eligible]).

Newly discovered patents and publications

First, when considering the statutory phrase “reasonably could have raised,” courts have differed on whether after-discovered patents and publications could not “reasonably” have been raised.

The most common scenario is where a patent challenger alleges that it conducted a diligent search and did not find a reference, but then had a good reason to find it later: for example, a third-party subpoena might turn up prior art that does not index precisely to relevant keywords but nonetheless bears on validity; or a publication might have only been available in a foreign language, and thus have required more time to find.

One influential opinion from Judge Roy Payne of the US District Court for the Eastern District of Texas all but guaranteed a context-sensitive approach by describing the issue as fact intensive. The Court explained that “[i]n congressional debates, one of the key architects of the AIA explained that ‘reasonably could have raised’ is meant to include prior art that a petitioner actually knew about or that ‘a skilled searcher conducting a diligent search reasonably could have been expected to discover.’”[2] The court added, “[t]he issue of whether a skilled, diligent search reasonably should have uncovered a reference is a question of fact.” Subsequently, courts have identified several types of “evidence that a skilled searcher's diligent search would have found the [newly raised reference].”[3] “One way to show what a skilled search would have found would be (1) to identify the search string and search source that would identify the allegedly unavailable prior art and (2) present evidence, likely expert testimony, why such a criterion would be part of a skilled searcher’s diligent search.” Id. One district court denied a motion for summary judgment regarding estoppel when the patent challenger showed it had hired two patent search firms and that neither found the new reference.

The law is still developing in this area, and because the diligence inquiry is a “question of fact,” patent challengers are able to identify any reasoning or rationale that may have persuasive value to the court – as well as producing expert evidence and testimony on the question.

Obviousness combinations with system or public use art

Courts have most sharply disagreed over when a piece of product prior art forms a different “ground” than a publication raised as a ground in an IPR. Consider, for example, a prior art product – such as a laptop – and a user manual for that laptop which describes every aspect of its functionality perfectly and completely. If an IPR could have been based on the product’s manual, and the manual’s teachings are substantively the same as the product, would it be an unfair circumvention of the statute to allow petitioners to cite the product when the manual was available (or perhaps, even expressly raised in the IPR petition)?

The majority of courts have said that the product art is, by definition, a different ground than the published manual because a product (or public use) is a different statutory type of prior art.[4] One decision notes that “the Federal Circuit repeatedly has characterized specific prior art references (or combinations thereof) as distinct ‘grounds’ that may be asserted in IPR proceedings . . . For these reasons, the Court interprets the term ‘ground’ in the IPR estoppel provision of 35 U.S.C. § 315(e) to mean a specific prior art reference or combination thereof,” exclusive of products and product combinations, since IPRs categorically cannot raise product-based prior art or combinations.[5] Such decisions have criticized contrary opinions as judicial policymaking based on an extra-statutory concern about unfair end-runs around the statute.[6]

Other courts, expressly concerned with purported unfairness to patentees resulting from an alleged end-run around the estoppel statute, have crafted a “superior and separate reference” standard, under which physical product prior art is subject to IPR estoppel unless it is not only different from but also substantively stronger than the patent or printed publication.[7] In essence, these courts require the product to be materially different from corresponding publications or patents, such that the product is truly a different “ground” of invalidity than corresponding writings.

Still other courts apply an apparently subjective standard which provides that IPR estoppel applies to physical products only where a party is attempting to “cloak” its reliance on a prior art ground based on a patent or printed publication that discloses all elements of the physical product or system.[8]

Among other considerations, courts attempting to distinguish true reliance on a product from reliance on a publication have considered whether the patent challenger cites nonpublic documents or source code as part of its characterization of the product prior art. For example, source code that is nonpublic but still operational in a public product may still qualify as prior art.

Accordingly, some courts have found that a product forms a different “ground” when the patent challenger’s characterization of the product “does not solely rely on publicly available documents” or when an invalidity argument “heavily relied on…confidential, internal documents.”[9] These decisions may also be motivated by a notion that nonpublic documents (such as source code or other confidential information that is most commonly produced only in response to a litigation subpoena) were not “reasonably” available to an IPR petitioner and thus could not “reasonably have been raised” in the IPR.

Some also implicitly embrace the “separate and superior” standard, recognizing that the additional disclosures imparted by confidential information differentiate the product from its public manual: for example, confidential documents designed for a product’s developers which characterize a product may provide more fulsome discussions and disclosures than a public user-facing manual.

Another area of dispute over the statutory term “ground” is whether product art combined with otherwise-estopped publications or patents forms a different “ground” than the publications or patents standing alone. Consider, for example, a manual for one product (eg, a laptop) and a separate product (eg, a cellphone). While the manual for the laptop would be a printed publication (and thus barred – as having been available to raise), an IPR petitioner could not have raised an obviousness argument based on the written laptop manual combined with the physical cellphone product. Typically, courts treat combinations as a different ground than each reference standing alone: “[A] subset of an obviousness combination constitutes a different combination. That is to say, obviousness combination A+B+C+D is not the same as the combination A+B.”[10] Indeed, the Federal Circuit has held – in an Administrative Procedure Act challenge to an IPR – that the Patent Trial and Appeal Board improperly raised a new “ground” when it invalidated a patent as anticipated based on a reference cited in the IPR petition, because the IPR petition raised only the reference as part of a combination of that reference and another.[11] Some courts have leaned into this reasoning and held that “nothing estops [a patent challenger] from raising invalidity arguments based on a combination of written and physical references.”[12]

At the same time, courts have been reticent to fully embrace this principle in the estoppel context, because a party could effectively relitigate its entire invalidity case – even prior art reference actually litigated in an IPR petition – if they were combined with a new product.

Conclusion

IPR estoppel is a rapidly developing area of law with vastly different practices around the country and even within individual districts. Practitioners would be well advised to investigate both the general practice in a given district and the practice before a given judge when crafting their IPR estoppel arguments.

Learn more about IPR estoppel by contacting either of the authors or your usual DLA Piper relationship attorney.

________________________________________
[1] California Inst. of Tech. v. Broadcom Ltd., 25 F.4th 976, 991 (Fed. Cir. 2022).
[2] GREE, Inc. v. Supercell Oy, No. 219CV00071JRGRSP, 2019 WL 5677511, at *4 (E.D. Tex. Oct. 30, 2019).
[3] M-I LLC v. FPUSA, LLC, No. 5:15-CV-406-DAE, 2020 WL 13413830, at *12 (W.D. Tex. Aug. 20, 2020).
[4] Chemours Co. FC, LLC v. Daikin Indus., Ltd., 2022 WL 2643517, at *2 (D. Del. July 8, 2022).
[5] Willis Electric Co v. Polygroup Macau Ltd., --- F. Supp. 3d ---, 2023 WL 112733 (D. Minn. Jan. 5, 2023).
[6] Chemours, 2022 WL 2643517, at *2 (“The statute at issue was the product of considered debate and careful thought. Congress could have broadened the categories of prior art on which IPR could be requested. Congress could have dictated that estoppel applies to products covered by the paper art underlying the IPR where the paper art discloses the same claim limitations as the product. But Congress did not do so. Adhering to well-accepted canons of construction, it is not for this Court to ignore Congress’s omission and create additional bases for estoppel.”); Willis, 2023 WL 112733, at *19 (“Moreover, the court in Wasica did not engage in a close analysis of the statutory text…For these reasons, the Court declines to follow the reasoning in Wasica”) (citing Wasica Fin. GmbH v. Schrader Int’l, Inc., 432 F. Supp. 3d 448, 454 (D. Del. 2020)).
[7] Wasica, 432 F. Supp. 3d at 454.
[8] Id.; Bos. Sci. Corp. v. Cook Grp. Inc., No. 117CV03448JRSMJD, 2023 WL 1452172, at *2 (S.D. Ind. Jan. 31, 2023) (expressing concern for merely “swapping labels” on publication and system art).
[9] CliniComp Int’l, Inc. v. Athenahealth, Inc., 2020 WL 7011768, at *1 (W.D. Tex. Oct. 28, 2020) (“rely solely” standard); Avanos Med. Sales, LLC v. Medtronic Sofamor Danek USA, Inc., 2021 WL 8693677, at *2 (W.D. Tenn. Oct. 8, 2021) (“heavily relies” standard).
[10] Finjan, Inc. v. Proofpoint, Inc., Case No. 13-cv-5808-HSG, 2015 WL 9460295 at *1-2 (N.D. Cal. Dec. 23, 2015).
[11] M & K Holdings, Inc. v. Samsung Elecs. Co., 985 F.3d 1376, 1386 (Fed. Cir. 2021) (“Here, unlike in Wasica, Samsung asserted obviousness as to claim 3, but did not include anticipation as an alternative theory of unpatentability. We therefore vacate the Board's decision as to claim 3.”).
[12] Microchip Tech. Inc. v. Aptiv Servs. US LLC, No. 1:17-CV-01194-JDW, 2020 WL 4335519, at *4 (D. Del. July 28, 2020).