Multiple IPR Petitions For Same Claim Are Often Not Repetitive

Post-grant review proceedings, including inter partes review, have become an important tool in managing intellectual property claims and risk. Refinements have been made since these proceedings first became available in 2012, including an updated U.S. Patent and Trademark Office trial practice guide in August 2018, the rule changes that implement the Phillips claim construction standard for IPRs, and the current practice of the PTAB consolidating petitions with common issues for efficiency purposes.^{1 2} However, relying on certain statistical examinations of IPR filings, some have expressed concerns that IPRs are unfair to patent owners and that the ability to file multiple petitions that address the same claim or set of claims of a patent permits "serial attacks" to "kill a valid patent" which is an "abuse" of the system.^{3 4}

To assess whether this strong rhetoric has any basis, it is useful to look at the underlying data. It is also useful to place the IPR process in historical context. IPRs have generally been understood to be most useful for complicated, unclear or vaguely worded claims that are more difficult for juries or courts to assess than the subject matter experts at the patent office. And, one purpose of IPR proceedings is to provide a mechanism for early adjudication of patent validity disputes before parties and courts expend significant time and resources in a district court case. To ensure early and efficient adjudication, rules have been put in place that restrict the timing and scope of any given petition for IPR. For example, petitions must be filed within one year of service of a complaint alleging infringement, they must contain no more than 14,000 words, and they must contain all of petitioners' arguments and evidence (incorporating argument by reference from an expert declaration, for example, is not permitted).⁵ In addition, if a patent survives IPR, the petitioner may have limited invalidity arguments in district court.

In combination, these restrictions force would-be petitioners to make a difficult risk assessment at the time of filing. At the point in the district court case that these petitions are being filed, there are many unknowns. For example, early in a case, it is not clear which patent claims a patent plaintiff will ultimately assert at trial; it is not uncommon for patent plaintiffs to nominally assert many claims from one or more patents early in a dispute, only to significantly drop the number of asserted claims later on, as the case develops. Also, it is often unclear early in a case what claim construction positions a patent plaintiff may settle on for purposes of infringement. Often, patent claims are long and technical, with multiple elements that may raise claim construction issues that need to be resolved. In order to successfully challenge a patent claim with uncertain scope, each opening petition must thoroughly address any and all issues potentially raised by a particular claim as well as typically address various conditional if/then scenarios. For these reasons, it is not uncommon for a petitioner to file multiple petitions that raise different arguments challenging a set of patent claims.

It is in this environment that some commentators have suggested that the practice of filing more than one petition directed to a particular patent claim is abusive and is also statutorily suspect in view of the estoppel provisions of 35 U.S.C. § 315. For example, one such commentator has argued that multiple petitions may be barred because once a petitioner obtains a "final written decision" for a petition for "review of a claim in a patent under this chapter," the petitioner "may not request or maintain a proceeding before the Office with respect to that claim."[[N:Supra, {4}.]] The conclusion is that these provisions at least implicitly prohibit the filing of more than one petition on a patent claim because if only a single final written decision is permitted, why would multiple petitions be permitted?

The full language of the statute does not support this argument. The full language of § 315 reads as follows: A petitioner or real-party-in-interest "may not request or maintain a proceeding before the Office with respect to that claim on any ground that petitioner raised or reasonably could have raised during that inter partes review." As discussed above, the subject matter requirements for the contents of a petition, in conjunction with strict word count limits and the uncertainty inherent in the early stages of litigation, leads directly to the conclusion that, if a petitioner has more than one meritorious argument directed to a common set of claims, it is often simply not feasible to put all such arguments in a single petition. Thus, multiple petitions are reasonably required and are not barred by 35 U.S.C. § 315.⁶

The more controversial argument in recent commentary is whether filing more than one petition against a particular claim, even if not barred, should be considered to be "repetitive" or abusive. One recent study asserted that Apple Inc. and Microsoft Corp. – which have filed many petitions for IPR (and been the subject of many patent lawsuits) – engaged in "duplicate petition practice" in over 50 percent of their IPR petitions, with a corresponding implication that this was evidence of abuse.⁷ Others, including USIJ, have cited to this study to argue that it is evidence of "serial attacks on high quality patents at the PTAB."⁸

We looked at the numbers. As an initial matter, the recent analysis does not provide any insight into the quality of patents challenged in IPR petitions one way or the other. Setting that aside, there is also little evidence of "serial attacks" and "duplicate filings." The incidence of such "duplicate" petitions appears to be substantially less than claimed, and where it does happen, it is often in situations where many claims of a single patent were challenged in reasonable circumstances, in view of the restrictions noted above.

First, in the underlying empirical analysis cited in the USIJ report, the authors overcount the number of duplicative petitions. The authors counted as "duplicative" every petition that includes one claim in common with one other petition by the same entity. But, this methodology overcounts the first petition. As an example, if Apple filed two petitions as to one set of claims, both of those petitions were counted as "duplicative," rather than just the second petition. Counting both petitions as duplicative misleadingly implies that Apple should never have filed any petition at all for these claims. When the data is adjusted for this type of overcounting, the percentage of "duplicative" petitions drops significantly.

Another way in which the authors overcount is by including petitions that were filed by multiple entities, as is common where the patent plaintiff filed suit against multiple defendants around the same time in district court. Where a patent owner sues six defendants, it is not unreasonable to have six petitions for IPR, assuming that each defendant should be able to use the IPR process as part of its defense. Counting multiple petitions from multiple parties in such situations fails to recognize that the patent plaintiff created the potential for duplication in the first instance by suing multiple parties.

For example, in one of the listed matters in the quantitative study, each of Apple, Samsung Electronics Co. Ltd. and Google Inc. is listed as either a petitioner or real-party-in-interest for the same two petitions filed against six claims of one patent. Rather than viewing this as an example of efficiency (three parties getting together to file only two IPRs), the article counts this as an instance of six duplicate petitions (two each by Apple, Samsung and Google). This overcounts duplication and implies that not all parties should be able to employ the IPR process. This type of practice does not suggest abuse and repetitive filings. Further, this alleged duplication ignores the fact that the PTAB often consolidates petitions that raise similar issues as to similar claims by the same parties, thus avoiding inefficiencies and militating against any "serial attack" issues that were an issue in such matters as General Plastics. Indeed, in a recent decision to institute, a PTAB panel noted that multiple petitions directed to the same or similar claims can be addressed efficiently by the board, would not waste board resources, and is appropriate in light of the potential for estoppel.⁹

Finally, as discussed above, the empirical analysis overlooked the procedural restrictions placed on petitioners as a cause for "duplicative" petitions. In contrast to the complaints filed by patent plaintiffs in district court, IPR petitioners must include in each petition its entire case in a limited set of words/pages – petitioners cannot change the claims, asserted prior art, or theories later in the proceedings. Thus, for long and complicated claims, or where a patent plaintiff has asserted a large number of claims in district court from a patent (a not-uncommon practice), it may be necessary to file multiple petitions to cover all of the issues raised. If the number of "duplicative" petitions in the authors; data is conservatively reduced to account for petitions including challenges to more than 15 claims, the percentage of "duplicative" petitions drops further.

The table below shows the percent of "duplicative" petitions adjusted for the foregoing issues:

Although it is true that abuse of the PTAB should be taken seriously, and efforts to place undue leverage on a patent owner by filing an unreasonable number of petitions against a patent owner should be discouraged, the current practice before the PTAB suggests that some repeat-filers (that are frequent defendants in district court) are practicing significant restraint in the issues they are bringing before the board.