Masters of Their Own Petition: The Federal Circuit’s Latest Stance on AAPA

by Dennis Crouch

One ongoing debate at the PTAB has been the role of Applicant Admitted Prior Art (AAPA) in inter pates review proceedings. AAPA refers to statements made by the patentee that admit that certain subject matter is part of the prior art. These statements are typically found in the background section and might include language like:

  • “It is well known that …”
  • “Conventional systems use …”
  • “Prior work includes …”
  • “As described in [reference], it is known that …”

Examiners rely heavily on AAPA, but IPR proceedings are limited to anticipation and obviousness grounds “and only on the basis of prior art consisting of patents or printed publications.”   In Qualcomm Inc. v. Apple Inc., 24 F.4th 1367 (Fed. Cir. 2022) (“Qualcomm I”), the Federal Circuit sided with the patentee (Qualcomm), holding that AAPA contained within the challenged patent does itself constitute such prior art.  Still,  the obviousness determination looks at evidence beyond the prior art, such as the level of skill in the art, and motivation to combine references.  And, the court held that AAPA could be used to prove those facts. The court noted particularly that a petitioner may properly rely on AAPA “as evidence of the background knowledge possessed by a person of ordinary skill in the art” for purposes such as “furnishing a motivation to combine” or “supplying a missing claim limitation” as part of the obviousness analysis.  But see Arendi v. Apple (Fed. Cir. 2016).

Following Qualcomm I, the USPTO issued updated guidance in on the treatment of AAPA in IPRs. Director Katherine Vidal’s new guidance explicitly acknowledged the Federal Circuit’s ruling that “under § 311, ‘patents or printed publications’ that form the ‘basis’ of a ground for inter partes review must themselves be prior art to the challenged patent” and not the challenged patent itself or any admissions therein. The guidance clarified that it is “impermissible for a petition to challenge a patent relying on solely AAPA without also relying on a prior art patent or printed publication.”

But, the updated USPTO guidance established what would later be termed an “in combination” rule, providing that “[i]f an IPR petition relies on admissions in combination with reliance on one or more prior art patents or printed publications, those admissions do not form ‘the basis’ of the ground and must be considered by the Board in its patentability analysis.” The guidance identified several permissible uses of AAPA in IPR proceedings, including: “(1) supply[ing] missing claim limitations that were generally known in the art prior to the invention…or the effective filing date of the claimed invention; (2) support[ing] a motivation to combine particular disclosures; or (3) demonstrat[ing] the knowledge of the ordinarily-skilled artisan at the time of the invention…for any other purpose related to patentability.”

Back to Qualcomm: On remand, the PTAB applied the USPTO’s “in combination” rule to conclude that Ground 2 of Apple’s IPR petition complied with § 311(b). The Board determined that since Apple’s challenge relied on AAPA “in combination with” prior art patents (Majcherczak and Matthews), the AAPA did not form “the basis” of the ground in violation of § 311(b). The Board rejected Qualcomm’s argument that Apple had expressly included AAPA as part of the “Basis” of Ground 2 in its petition tables, concluding that these statements merely described how the grounds included both AAPA and prior art patents used “in combination.”

On appeal, the Federal Circuit has once again reversed. Qualcomm Incorporated v. Apple Inc., Nos. 2023-1208, 2023-1209 (Fed. Cir. Apr. 23, 2025) (Qualcomm II). The new panel concluded that Board’s interpretation contravened the plain meaning of § 311(b). Rather, the Federal Circuit held that Apple’s express designation of AAPA in the “Basis” of Ground 2 was determinative, ruling that the Board erred in concluding the challenged claims were unpatentable under Ground 2 when that ground improperly included AAPA as part of its basis in violation of § 311(b).

Importantly, the court distinguished between permissible “reliance” on AAPA (such as to indicate general knowledge of a person of ordinary skill) and impermissible inclusion of AAPA in “the basis” of a ground. In this case, it was easy because Apple’s petition expressly designated the AAPA as a “Basis.” The court explained those express statements were dispositive since the “IPR petitioner is the ‘master of its own petition.'” quoting Intuitive Surgical, Inc. v. Ethicon LLC, 25 F.4th 1035, 1041 (Fed. Cir. 2022).  Thus, this result here may be a primarily procedural hurdle that petitioners can overcome by simply recharacterizing their use of AAPA.

Appealability: The case also involved an important question about whether the 311(b) issue is appealable because it focuses on institution questions.

35 U.S.C. § 314(d) bars review of “ordinary” challenges to the “threshold” institution decision.  But here, the court concluded that Qualcomm’s appeal focused on appropriate questions regarding “the manner in which the agency’s review proceeds once instituted” rather than non appealable issues of “whether the agency should have instituted review at all.

This case thus intersects with Acting Director Stewart’s March 2025 implementation of of new enhanced discretionary denial review of IPRs centralized within her office.  While 35 U.S.C. § 314(a) provides the Director with explicit power to determine “whether to institute an inter partes review,” Qualcomm clarifies that this discretion cannot extend to considering grounds that violate the statutory limits of § 311(b).  The result here really is something of an imbalanced rule where petitioners are unlikely to be able to appeal any issue from a denied petition, but patentees will still often have an opportunity if the ultimately lose the final written decision.

= = =

I just want to note an aside here regarding permissible and impermissible uses of AAPA.  The starting point for the discussion is the longstanding use of AAPA as an admission of prior art that serves as an extraordinarily powerful tool during both examination and litigation. But, for prudential reasons, we have an artificial limitation on the types of prior art that can be used during IPR petitions – and general admissions do not qualify.  But, this is a weird setup: they admitted its prior art, but we can’t use it as prior art.  In my view, this then drove the PTAB to bend backward to permit its use.

Along these lines, I also wanted to comment on the court’s indication that non-prior art evidence can be used to supply missing limitations as part of the obviousness analysis.  There are a couple of key Federal Circuit cases on point.

  • Koninklijke Philips N.V. v. Google LLC, 948 F.3d 1330 (Fed. Cir. 2020); and
  • Arendi S.A.R.L. v. Apple Inc., 832 F.3d 1355 (Fed. Cir. 2016).

In Philips, the court affirmed that the PTAB can use a skilled artisan’s general knowledge about “pipelining” to fill a missing limitation found in the claims but not in the asserted prior art reference.  This result stands in contrast to Arendi where the court rejected the Board’s “common sense” rationale for adding a search functionality to Pandit’s “Add to address book” feature. The critical differences? In Arendi, the missing limitation was central to the invention (not “peripheral”), the Board relied solely on conclusory statements without reasoned explanation, and the technology wasn’t “unusually simple.” These cases establish that while common knowledge can supply missing limitations, such reasoning requires substantial evidentiary support, especially when the missing element is fundamental to the claimed invention.​

Taking these in chron order: Arendi established important guardrails – requiring substantial evidentiary support when using common knowledge to supply a missing limitation, particularly when that limitation is central to the claimed invention. Philips then clarified that this doesn’t categorically prevent such reasoning, so long as it’s properly supported by evidence that the knowledge in question was indeed within the skill set of ordinary artisans in the field.