A petitioner in an inter partes review may request to cancel as unpatentable one or more claims only on a ground that could be raised under section 102 or 103 and only on the basis of prior art consisting of patents or printed publications.  This is expressly set forth in the statute, 35 U.S.C. § 311(b).  Notwithstanding this limitation, there are creative ways to bring other challenges, such as written description challenges under 35 U.S.C. § 112, ¶ 1, before the Patent Trial and Appeal Board (“PTAB”).  And there are good reasons for doing so.  Notably, PTAB judges, unlike district court judges that would otherwise preside over § 112, ¶ 1 arguments, have technical expertise—and in some cases, prior patent prosecution experience—and, consequently, are typically more receptive to hyper-technical arguments.

Written description attacks against means-plus-function (“MPF”) claims are precisely the sort of technical arguments that one might expect to be more favorably received by the PTAB than by a district court judge.  See 35 U.S.C. § 112(f) (previously named, and still commonly referred to as, § 112, ¶ 6).  Although § 311(b) appears on its face to bar such challenges in an IPR, the Federal Circuit’s recent decision in Uniloc USA, Inc. v. Sega of America, Inc. et al., No. 2016-2000 (Fed. Cir. Oct. 23, 2017) appears to open the door to a novel tactic for bringing such challenges.

In Uniloc, the Petitioners (Sega et al.) had challenged the claims of U.S. Patent No. 5,490,216 (“the ’216 Patent”) as being anticipated by U.S. Patent No. 5,509,070 (“Schull”). Because Schull had an effective prior art date later than the provisional applications to which the ’216 Patent claimed priority, the Petitioners’ anticipation theory could only prevail if the Petitioners were successful in defeating the priority claim to the provisional applications. The Petitioners did not, however, present a conventional § 112, ¶ 1 argument that the claim elements were not disclosed in the provisional applications. Instead, the Petitioners argued that a claim recitation calling for a “unique ID generating means” was properly interpreted as a MPF limitation under § 112, ¶ 6, and that the provisional applications merely disclosed the claimed function and did not describe the structure (i.e., an algorithm) for performing that function. The PTAB and the Federal Circuit agreed with the Petitioners’ analysis, and held that the priority claim was ineffective because the priority documents did not comply with the requirements of § 112, ¶ 6.

In Uniloc, the Petitioners were ultimately advancing a § 102 challenge based upon a third-party prior art reference. Uniloc therefore dealt with a challenge that clearly fell within the permitted scope of § 311(b). Even so, the Federal Circuit’s holding appears to open the door to presenting § 112, ¶ 6 attacks per se against a substantial fraction of challenged patents.

Here’s how this could work.  Say a petitioner wishes to challenge a patent (“Child”) that claims priority via continuation to an earlier-filed application (“Parent”) that was published more than one year prior to the date upon which the Child was filed.  Because the first patent application in a family is typically published 18 months after filing, and because examination often takes two years or more to complete, it is quite common for continuation applications to be filed well after the priority document has been published.  Imagine further that the claims in the Child contain a limitation that could reasonably be construed as an MPF limitation—say it recites a ‘means,’ a ‘module,’ a ‘unit,’ etc.—and that the specification merely states the function that the ‘means’ performs but does not describe the specific structures or algorithms for providing the claimed function.  This too is a common scenario.

Applying the holding in Uniloc, it appears that the Child is not entitled to the filing date of the Parent because the Parent specification does not describe the structures or algorithms for performing the function recited in the claims of the Child.  Note that it seems to be irrelevant that the specification of the Parent may be identical to that of the Child—unless the Parent specification adequately describes the subject matter claimed in the Child, the priority claim is defeated.  Moreover, since the Parent was published more than a year before the Child was filed, the Parent is prior art to the Child.

Since the Child is a continuation of the Parent, the Parent—which is now prior art—likely discloses every element in the challenged claims except the structure required by the MPF limitation. As to this sole missing element, the Parent further describes the need to provide an element that performs this recited function. Thus, if the petitioner is able to leverage § 112, ¶ 6 (or § 112(f)) to defeat the priority claim, the petitioner will be rewarded with a nearly-anticipatory reference and an obviousness rationale for supplying the only missing element.

Although subtle, this tactic has potential to be highly effective against a substantial fraction of challenged patents—particularly against software patents which often claim modules for performing functions without specifically describing what those modules are or how they operate.  Although these patents could, in principle, be invalidated via direct § 112 challenge in a district court, petitioners may see substantial advantage to be gained by presenting these hyper-technical arguments before the trained patent specialists at the PTAB.