Patents Post Grant Blog

Inter partes reexamination, has created new opportunities for the public to challenge the patentability of some or all claims of issued patents filed on or after November 29, 1999 and to actively participate at each stage in the reexamination proceedings.  Indeed, as noted yesterday, inter partes filings, especially those relating to copending district court litigation,  have surged over the past several years.  Yet, this new proceeding has become a lightning rod of criticism for how it is managed in the USPTO.  Whether by academic rant, or judicial admonishment, inter partes reexamination is largely ridiculed as ineffective and open-ended in time, taking many times on the order of 7 years to conclude.  Yet, the delay in concluding an inter partes reexamination is impacted by many factors, many of which are outside of the control of the USPTO. 

For example, the inter partes reexamination statute burdens a requesting party with what some consider to be a significant estoppel provision.  Specifically, section 315(c) of title 35 of the United States Code provides that a third-party requester is estopped from asserting at a later time, in a civil action under 28 U.S.C. § 1338, the invalidity of any claim “finally determined to be valid and patentable on any ground which the third-party requester raised or could have raised during the inter partes reexamination proceedings.”(emphasis added).

As the vast majority of patents involved in inter partes reexamination proceedings are involved in concurrent litigation, requesters realize that the statute provides a “use it or lose it” option with respect to available prior art.  Therefore, inter partes reexamination proceedings, unlike ex parte reexamination proceedings which have no such statutory estoppel provision, are far more likely to include a greater number of substantial new questions of patentability (“SNQ”). 

In other words, if a requester is aware of a single primary reference (A), and 20 interchangeable, cumulative secondary references all showing the same technical feature, the requester is practically obligated to present all 20 SNQs.  As such, it is not uncommon for an inter partes reexamination proceeding to include on the order of 20+ proposed grounds of rejection, all of which are virtually cumulative with respect to the other.  Furthermore, in a concurrent litigation, it is much more impressive to inform the court that a claim is rejected on 20 different grounds than say, a single ground.  Thus, while the “stacking” of cumulative SNQs is especially problematic in inter partes reexamination, there also exists motivation for the same overkill in ex parte reexamination proceedings concurrent with litigation.

As the need to actively consider all SNQs in some form or another falls on the Office, the burden being twice as oppressive in inter partes reexamination since both parties participate, it is easy to see how inter partes proceedings can slow to a crawl.

Indeed, in one recent inter partes reexamination proceeding the Examiner lamented:

Currently, Congress has proposed removal of the “could have been raised” language from the statute in the most recent version of the patent reform bills.  Such a change would be helpful in eliminating the estoppel liability for inter partes reexamination.  Yet, this legislation appears hopelessly stalled with Congress distracted with health care issues.

So, what can and should the USPTO do about SNQ stacking now?

MPEP § 2240 precludes the granting of a second reexamination request while a first inter partes reexamination proceeding is pending, if the second request presents the same SNQ as the first.  In other words, the Office will refuse to examine the same SNQ twice, simultaneously.  A proper SNQ analysis includes determining whether or not art is cumulative with respect to art previously considered.  MPEP 2240 alters this analysis slightly to include art currently being considered.  Why not consistently apply this analysis to SNQs of a single request?  

Stated differently, the stacking of SNQs as outlined above is nothing more that the presentation of cumulative SNQs in a single request; why should the USPTO examine cumulative SNQs of a single request, where if divided across multiple reexamination requests, such SNQs would be denied?

While one simple solution to SNQ stacking may be to simply deny the cumulative under a more rigid interpretation of the current guidelines, however, with the significant estoppel effect of inter partes reexamination in mind, perhaps application of a stringent view of cumulative art would have a chilling effect on filings. 

Instead, I propose that the Office revise the inter partes rules to effectively control the number of SNQs considered during inter partes reexamination. 

In operation, the rules would provide that after identifying all viable SNQs of a request, those that are arguably cumulative with respect to other SNQs would be grouped together.  Once grouped, the SNQs are presented to the requester for the purpose of electing a single, representative SNQ for reexamination on the merits, non-elected, cumulative SNQs being held in abeyance until and unless needed, if rejections based on the representative SNQ are overcome. 

Under such a system, the requester would have the right to petition to substitute a non-elected SNQ if the elected SNQ of record is overcome.  The petition would require a showing that the held in abeyance SNQ meets the claims limitations relied upon by the Office which overcome the elected SNQ.  For example, one could imagine a situation in which it is discovered during prosecution that the secondary reference of the elected SNQ teaches away from the applied combination of references while a non-applied reference does not provide such a teaching away.  If the elected SNQ is not distinguished, there would be no right of petition.

Upon issuance of an Action Closing Prosecution (ACP), the Examiner would indicate that all rejections have been considered for appeal purposes and that the non-elected SNQs are believed to stand or fall together with the elected SNQ.  Should the requestor choose, for a significant additional fee, the SNQs held in abeyance would become available for briefing during the appeal.  The examiner would not be obligated to address the cumulative SNQs, but would be permitted to do so at his/her option in the Examiner’s Answer. 

It is envisioned that the Office would require a substantial fee for the ability of the requestor to brief non-elected SNQs to offset the potential of an increased level of effort at the Board of Patent Appeals & Interferences (BPAI).  The level of the fee should increase proportionately to the number of non-elected SNQs desired for briefing to discourage abuse at the time of appeal.

If the requestor does not opt to have the non-elected SNQs considered, the Board would not review these SNQs should the elected SNQ be affirmed. Of course, if the elected SNQ is reversible, the non-elected SNQs would then be considered, as enlightened by any requester/patent holder briefing.

In this way, examiner review time would be significantly decreased by the omission of prior art stacking, and requesters would not be faced with estoppel on the non-elected rejections (unless they truly are cumulative, and inferentially distinguished during the reexamination proceedings).  Likewise, Patent Holders lose nothing as the substitution of a new rejection, if not cumulative with one previously distinguished, would be appropriate where the rejection of record has been overcome.

The above recommendations are believed to be a positive step toward streamlining the current de facto requirement that all possible grounds of rejection regardless of their redundancy be actively considered throughout every inter partes reexamination proceeding to avoid estoppel.  Likewise, as a sizeable portion of inter partes petition practice and informal filing delays relate to the exceeding of page limitations, the de-stacking of SNQs will enable participants to more readily meet page limitations, further enhancing Office efficiency.