USPTO inter partes proceedings are not healthy for patents. Inter partes reexamination has resulted in the cancellation of all claims of a patent nearly 45% of the time.2 The probability of any single claim being canceled is likely above 70%. The primary appellant from the Examiner's decision in inter partes reexamination has been the Patent Owner nearly three times as often as it has been the Challenger.3

The success rate of Challengers in inter partes proceedings could be caused by a number of things, including the lower burden of proof before the PTO, the broader claim construction, and tendency of the higher-placed PTO officials to pay little deference to the decision of junior examiners. Whatever the true causes, however, it is noticeably easier to defend patent rights before a jury than it is to recover them from inter partes PTO proceedings, which is why accused infringers often make use of these proceedings.

This article is about constructing a winning case as a Patent Owner in an inter partes proceeding. And in particular, it is about how to win on the merits. Much has been said about inter partes procedure, probably because the procedure is at least partly ascertainable from the rules, even without actual experience in using them. The plethora of literature already available on inter partes review, despite the fact that (as of today's date) no one has ever filed one, bears witness to this.  

Yet little has been said publicly about the merits case. The merits case is not as easy to derive from published rules. Rather, it will vary according to the patent. So to say anything beyond a numbing recitation of the statute and the Graham factors has been difficult.

Which strategies have Patent Owners pursued that have resulted in success in inter partes proceedings? The following sections include my observations from inter partes reexamination practice.

If you are reading this because you are trying to decide how to approach an actual case, then let me first provide some free advice: get started now. You may have received a copy of a recent reexamination request or inter partes review petition from your prosecution counsel. This means that you have, worst case, approximately four and a half months (inter partes reexamination) or eight and a half months (inter partes review) to put together your case-in-chief. Like as not, you will need all of it.

I. Anticipation

The merits case in inter partes reexamination and inter partes review is usually about anticipation and obviousness.4 Where anticipation is put forward, the options are relatively limited. The first thing to do, of course, is to check whether the references actually teach the claim elements. There is often significant (and understandable) pressure from litigation counsel to make a § 102 challenge out of what should be a § 103 case. Often, the Challenger will opt not to submit a backup § 103 position. If this happens: good. If the missing limitation is technically significant: great. You will probably win this issue. Do not relax at this point, however, because it is within the Examiner's discretion in reexamination to fix this challenge by coming up with an alternate § 103 position.5 Try to imagine what that § 103 case could be, and proceed to section II, below.

Another common failing of anticipation cases is the arrangement of claim limitations within a reference. Some references put forward as § 102 references teach the claim limitations scattered throughout different embodiments, sometimes in ways that make a combination of the embodiments implausible. This should also be checked first, although it will not lead to victory as surely as a reference that is entirely missing a claim element.

If the Challenger has a plausible position for each claim element arranged as in the claim, then you have a number of possibilities. Let me speak about amendments first, even though amendment will not be the first option in many cases. A few years ago, I thought that an outright amendment to the claims was a mistake in reexamination, because a host of bad things happen when a claim is substantially amended. A substantial amendment wipes out past damages and raises certain intervening rights.6 If there is parallel litigation, some judges will take an amendment as a tacit admission of invalidity, which can affect a number of discretionary decisions made in the course of a case, such as a decision on a motion to stay or a motion for preliminary or permanent injunction. So in my view, it was always better to introduce narrowed subject matter as a new claim, and then argue the original claim in parallel with the new claim. However, in some cases, there is no parallel litigation, and past damages are irrelevant. In such cases, amendments (as opposed to the parallel introduction of new claims) have a significant psychological effect. This psychological effect is based on the fact that, without challenging the decision maker's original findings, an amendment can force actual reconsideration of an issue. Without amendment, there is a risk that counter-arguments presented after an initial decision has already made will be prejudiced, in the literal sense of that word. No one likes to be proven wrong.

In my experience, the psychological advantage of an amendment can be preserved even if an independent claim is amended to incorporate the subject matter of a rejected claim directly dependent from it. This is significant, because the independent claim thus amended does not lose past damages or raise intervening rights. So if it is possible to amend a claim in this way (e.g. if the dependent claim is asserted in parallel litigation), it is usually the first kind of amendment to consider.

After you have considered your amendment strategy, check whether the prior art date of the reference is sufficiently proven. The issue date of a patent is solid, but the § 102(e) date may not be, especially if it depends on a benefit claim. Check any underlying patents for disclosure issues and check the § 120 and § 119(e) claims for compliance with the statute—noting intervening statutory amendments and making sure to apply the correct law at the time the claim was made. For printed publications, the PTO may initially accept a date printed on a reference, including a copyright date. However, these dates are sometimes imprecise, and in close cases it is worth checking for the date of the first event that would actually meet the requirements of "publication". Be careful not to let the Challenger assert that the date of a conference is the date of publication of the conference proceedings, or that company-internal documents were necessarily "published" at all.

Next, check whether it is possible to swear behind the anticipating prior art. If it is, begin building that case now. Declarations under 37 C.F.R. § 1.131 are (properly) subject to much more scrutiny in inter partes proceedings than in application prosecution, and the successful ones are the exception. Someone who really knows the law of invention (conception, diligence and reduction to practice) will need to take charge of this task.

The next thing to look at is whether the anticipation case rests on one of two common foundational elements that can sometimes be weaker than other parts of the case: inherency and incorporation-by-reference. Inherency is a somewhat limited doctrine, and the PTO sometimes interprets inherency strictly. Nevertheless, the same forces that push Challengers to make § 103 cases into § 102 cases can also lead them into ill-advised applications of inherency. Incorporationby- reference doctrine is also misunderstood with some frequency by Challengers.

One can also argue that the reference does not enable the claims. This is best done with expert testimony, supported by explanation. If testing can be done to show the lack of enablement, so much the better, although the disclosure of the prior art is, ironically, sometimes so poor that it cannot be tested. If a case of nonenablement can be made over a reasonable scope of the claim, it has a good chance of success. However, it is time-consuming to construct such a case, especially if testing is used.

If a good case can be made using any of the above approaches, the chances of success on anticipation are relatively high. If such a case cannot be made, then you are left with an approach that tends to be less successful: arguing that the claims should be more narrowly construed than the Challenger advocates. This tends to be a difficult argument to win because the PTO follows the "broadest reasonable interpretation consistent with the specification" for non-expired patents.7 So in order to limit the claim language, the Patent Owner will likely need to make a convincing argument based on the specification. The PTO tends not to be swayed by narrow court constructions that add limitations to the literal language of the claims (as opposed to explaining the meaning of certain terms).

II. Obviousness

If the challenge is one of obviousness, then just about any of the techniques described for anticipation can be used, with the caveat that the analysis of missing claim limitations should take the entire combination into account. With obviousness, however, one also adds the possibility of arguing that there is no reason to combine or modify references, or that the objective evidence shows that the reason was not sufficient.

Despite these additional battlefronts in a § 103 challenge, the author's experience has been that § 103 challenges are more difficult to overcome than § 102 challenges. This is primarily because of the tendency, noted above, for Challengers to stretch their § 102 case. A § 103 challenge grants the PTO greater leeway to remedy flaws or omissions in the prior art.

The issue of obviousness can also be asymmetric. This simply means that it is easier for one side to make its case than the other. Such issues occur frequently in patent law: for example, if a Patent Owner finds a careless email by an infringer praising the value of the invention in the infringer's product, this can have a significant impact on the damages case. If the infringer believes that the invention is not, in truth, driving the entire value of the product, the infringer now has to show numerous other techniques that also contribute to the value, in order to dilute the effect of a single email.

Obviousness can be like this. The initial case of the Challenger is relatively easy to put together, but a complete rebuttal by the Patent Owner is hard. The Challenger must put forward a credible case, based on the prior art, for the obviousness of all elements of the claims. The Challenger must also provide reasons to support that obviousness (e.g. a motivation to combine multiple references). This is usually done with a selected few references and perhaps expert testimony.

The Patent Owner can also make arguments about the content of these references, and can adapt any of the approaches under section I, above. This is only the "easy" part of the case. The "hard" part involves building an evidentiary case beyond the references at hand, to allow the factfinder to "look at the situation before and after it appears".8 This can be done, for example, by finding prior art that teaches away from the invention, by documenting in the post-art the reaction of industry to the invention, or by establishing more traditional objective indicia of non-obviousness such as commercial success. The hard part of the case follows the guidance of the Federal Circuit in some recent panel decisions, reasoning that:

Courts, made up of laymen as they must be, are likely either to underrate, or to overrate, the difficulties in making new and profitable discoveries in fields with which they cannot be familiar; and, so far as it is available, they had best appraise the originality involved by the circumstances which preceded, attended and succeeded the appearance of the invention.9

Patent Owners stop with the "easy" case with great frequency. The reasons for this can be good or bad. In some cases, cost is a limiting factor. Doing investigation beyond the references accounts for more than half of the cost of a full rebuttal, and may simply not be feasible. In some cases, even good inventions will not have evidence of events that "preceded, attended and succeeded the appearance of the invention", where the evidence is probative of the question of obviousness. Yet in some cases, the hard part of the case is left undone because Patent Owners simply do not know how to put it together. This can be caused by application prosecution training—where cost is almost always a primary driver of strategy—or by a simple lack of experience building a more complex case for non-obviousness.

Tracing the history of the field before and after the invention is hard to do, but often worth it. After all, the art initially before the PTO was selected by the Challenger—why should we limit the PTO's consideration to that selection? There is some risk, of course, that prior art will be discovered that is better than the Challenger's. In most cases that have reached the level of a PTO inter partes proceeding, however, the prior art has been searched by one or more patent office and intensively scrutinized by accused infringers. The art before the PTO is fairly likely to be the best art available.

What is there to look for in the literature that justifies the risk of finding better prior art? There is a seemingly intuitive notion among patent attorneys that the more art there is of record, the more obvious the invention will seem. I have found the opposite to be true. In most cases, the factfinders (Examiners and PTAB judges) are not, except by fortunate coincidence, persons of ordinary skill in the art. Rather, they must be sufficiently trained to be able to recreate the view of a person of ordinary skill in a wide variety of technical fields at a wide variety of times. They do this in the initial instance by reading the patent and the art provided by the Challenger. This can lead to an overly simplistic view of the art. Where a full picture of the art is presented, it sometimes reveals a complexity that makes the invention less obvious.

There are a few questions to ask during the development of this evidence. It is usually easiest to start by asking the inventors why they thought the invention was worthy of filing a patent. This often leads to a discussion of the problem to be solved, and a discussion of how other people were trying to solve the problem. Documenting alternative approaches to the same problem might show that the invention was not the next logical step in the evolution of technology. It may also reveal beliefs in the art that would have led a person of ordinary skill away from the invention. Review articles and textbooks from the relevant time might be examined. If they do not discuss the invention, that might be useful evidence. If they reveal predictions of the future that do not involve the invention or disparage techniques used in the invention, that is even better.

The post-art is also worth investigating. Sometimes, it is possible to document gradual adoption of the invention that occurs after publication of the patent, the publication of a related journal article, or distribution of a product embodying the patent. In some cases, review articles trace the post-publication history of the invention, and in some cases, they may even reveal previously unknown praise of the invention.

I have found it surprisingly productive to dig deeper into the prior- and postart. At a minimum, a broader and often more complex picture of the art usually comes to the fore, which can be useful for a non-obviousness case. With surprising frequency, one can also locate game-changing evidence. Here are a few examples of things from the last few years of my practice that we have unexpectedly located during the course of such investigation:

  • a third-party article attributing the Patent Owner's recent commercial success to a critical feature of the claimed invention;
  • a product brochure of the Challenger, after the issue date of the patent, calling a critical feature of the claims "innovative" when used in the Challenger's own products;
  • a prior art patent application of the Challenger disparaging the technique of the invention, and recommending a different approach;
  • a third-party technical article calling the results of the invention "surprising";
  • an article by the Challenger's litigation expert, calling the invention an "important discovery" and saying that the discovery was first made by the named inventors.

You will not find these sorts of things in every case, of course, but with persistence, you will find them in more cases than you initially expect.

Some Patent Owners may fail to develop objective evidence, believing that the PTO places less emphasis on the "hard" part of the case. I agree with this belief on average. However, it is less true before the PTAB than before Examiners, and more-or-less completely incorrect before the Federal Circuit. I think this has to do with the fact that most reexamination Examiners have seen numerous poorly constructed cases for commercial success, and for that reason tend to de-emphasize all forms of objective evidence. In contrast the Federal Circuit judges see more cases from litigation with fully developed records, and generally only see cases that are worth appealing. Thus, as one goes up the appeals ladder, the paradigm shifts. This shifting paradigm has caused many a Challenger to watch in dismay as an initial rejection of the claims crumbles at higher levels of appeal.

The latter consideration implicates a bit of procedure as well. Before the Federal Circuit, the PTAB's ultimate conclusion of obviousness is reviewed de novo, whereas the underlying facts are reviewed for substantial evidence. A Patent Owner who does a good job developing objective evidence should push for factual findings on that evidence, even if the ultimate conclusion of the PTO is that the claims are obvious. Before the Federal Circuit, there is a significant difference between a PTAB rejection that finds the Patent Owner's objective evidence credible, but insufficient, and a PTAB rejection that simply finds the Patent Owner's evidence not credible.

Patents can be recovered from inter partes proceedings, while maintaining a reasonable claim scope. To maximize the chances of such recovery, Patent Owners need to be cognizant of their options. The best cases require time to develop, meaning that Patent Owners should be prepared to pursue aggressively all available options from the moment the inter partes challenge is received.