Partial-Institution Decisions Blessed by En Banc Federal Circuit

SAS v ComplementSoft (Fed. Cir. 2016)

Today, the Federal Circuit denied SAS’s en banc request challenging the USPTO’s approach to partial-institution of inter partes review petitions.  In a substantial number of cases, the PTO only partially agrees with the IPR petition and thus grants a trial on only some of the challenged claims.  In the present case, for instance, SAS’s IPR Petition challenged all of the claims (1-16) found in ComplementSoft’s Patent No. 7,110,936, but the Director (via the Board) instituted review only on claims 1 and 3-10.

The statute seems to side with SAS: The Board “shall issue a final written decision with respect to the patentability of any patent claim challenged by the petitioner” 35 U.S.C. § 318(a).  However, the appellate panel in this case (following prior precedent) held that “Section 318(a) only requires the Board to address claims as to which review was granted.”

In its petition, SAS wrotes:

Because § 318(a) is clear and unambiguous in requiring a final written decision as to “any patent claim challenged by the petitioner,” the PTO had no authority to adopt a contrary rule authorizing IPRs “to proceed on all or some of the challenged claims,” 37 C.F.R. § 42.108(a). Regardless of efficiency or workload concerns, the PTO’s rulemaking authority “does not include a power to revise clear statutory terms.” Utility Air Regulatory Grp. v. Environmental Protection Agency, 134 S. Ct. 2427, 2446 (2014).

In what appears to be a 10-1 decision, the Federal Circuit has denied SAS’s petition for en banc review.  Although the majority offered no opinion, Judge Newman did offer her dissent (as she did in the original panel decision).




49 thoughts on “Partial-Institution Decisions Blessed by En Banc Federal Circuit

  1. 4

    One problem with Newman’s take here is that there is no estoppel for the challengers of granted patents as to their own selection of claims to challenge or not – thus leading against the very logic Newman wants to rest upon.

  2. 3

    patent bras I have a question for you guys that might be well up your alley. Today let’s hypothesize I’m tots not looking at an app that has a device that, theoretically (and I stress theoretically) emits a certain gas. Like let’s just pretend it is a steel doorstop that emits oxygen at a certain measurable rate after it is formed until the oxygen is gone out of it. But, let us further hypothesize that the inventor themselves seem to be jumping through theoretical hoops all throughout the lengthy theoretical application thinking that their “theories”, “models” and “calculations” show that a steel doorstop that does this emission could possibly happen (but they don’t actually say for sure). What does patentlyo say the lawl says about such theoretical devices? Enablement problems or no? Just as a heads up this doesn’t seem like a super easy thing to make.

    Further let’s hypothesize that the applicant claimed this device that emits oxygen as a device with a “wherein” tacked on the end that says “wherein a released gas of the doorstop device observed as oxygen with a measuring machine is x/unit volume”. Would that seem to not be improper English at best as the gas is not technically in the device (as it would be outside the device)? Might this not also be seen as an attempt to mix statutory categories (Device with composition of matter) as the gas is not actually part of the device yet is in the claim?

    1. 3.1

      These problems are largely due to the fact that until recently the PTO had discouraged examiners from giving 112 enablement rejections even when well deserved, plus the Fed. Cir. and D.C.’s make challenging 112 enablement by summary judgment in litigation very difficult. Nor can this be raised in an IPR, except for amended claims.

      1. 3.1.1

        I keep in hearing of this “the Office discourages giving 112 rejections” (mostly on these boards).

        Has anyone produced anything of note to substantiate this directive?

        It would seem that such a callous disregard for one’s duty to examine fully would have been made “an issue” by at least one diligent examiner at some point. It cannot be that examiners are universally such indolent sheep, can it be?

  3. 2

    I believe that Anon is correct that there is no estoppel pursuant to the statute. MM, you do, however, raise a potentially interesting estoppel argument at the district court (post IPR). Or maybe a summary judgment motion arguing that the dependent claim(s) cannot possibly be invalid over the prior art since the PTAB confirmed the validity of the broader independent claim. And this is perhaps one of the few times where the PTAB’s broader standard of review helps a patent holder: In other words, if the independent claim was valid under the BRI standard, then certainly the dependent claims must be valid using the narrower district court standard.

    Anon: What say thee?

    1. 2.1

      I can believe that estoppel does not attach to dependent claims on which review was not granted. On the other hand, I would note that both the CAFC and the SCotUS have adopted the view in Cuozzo that dependent and independent claims are implicitly tied together, and that even when the petition does not address one or the other, it might still be implicated. In other words, I would not be shocked if a future case—squarely presented with the question—were to hold that a finding of no-unpatentability for an independent claim implicitly shielded a dependent claim from attack on the same grounds that failed for its parent.

      In other words, we really need to wait for a ruling on this question before we can pretend to know the answer. It is not 100% clear merely from reading the statute or the already-existing precedents.

      1. 2.1.2

        In other words, I would not be shocked if a future case—squarely presented with the question—were to hold that a finding of no-unpatentability for an independent claim implicitly shielded a dependent claim from attack on the same grounds that failed for its parent.

        I’ll go further and say that I would be surprised if, under the particular facts of the hypothetical I presented, a judge let the argument go forward.

        If you’re estopped from arguing that reference A renders X+Y obvious (according to the IPR statute), basic logic and fairness (not to mention judicial economy) says that you’re also estopped from arguing that reference A + reference B renders X + Y + Z obvious if all that B brings to the table is a teaching that “adding Z is good”. The point being that you already made — and lost! — the argument before PTAB. X + Y was deemed non-obvious and an additional limitation can’t possible change that ruling.

        Consider how it might play out otherwise. Defendant performs steps X + Y + Z (i.e., the steps of the dependent claim). But rather than challenge that claim in IPR, it chooses instead to challenge just the independent claim (X + Y) in the IPR (i.e., the independent claim). If it wins, awesome! If it loses and it’s not estopped, it gets another bite at the apple in the district court relying on the same art and the same argument. Sure, it needs to address “Z” but — according to the facts here — Z adds nothing to the patentability of the claim. That’s just not right (again, I’m accepting the estoppel provisions of the statute for the sake of the argument).

    2. 2.2

      You still need to look at the depdendent claim on its own (due in part because you are bringing more than just the prior art that the independent claim had had brought against it).

    3. 2.3

      But IPR-allowed claims can still be attacked in D.C. litigation on any ground that could not be raised in an IPR, i.e., anything other than prior patents or publications.

      1. 2.3.1

        Another point Paul – the “at least not 102/103 prior art” is a half measure at best, and certainly not one that would be stopped at in any fight over a valuable area covered by patent rights.

        More and more, what Congress did with IPRs seems half-baked.


          As often occurs, I am not sure what you meant to express. But it was NOT an accident that IPRs were limited to only PART of even 102 and 103 defenses – those based only on patents or publications. It was argued that this PTO proceeding is not well suited to get into controlling and deciding complex discovery and fact issue disputes over things like prior public use and on sale bars, prior products contents and operations, etc., and decide them in one year.


            Yes Paul, thank you for adding the particulars of that portion of any (fuller scale) fight that would likely ensue over valuable and/or contested items.

            The very fact that IPRs were aimed at satisfying less than a complete matter makes it rather apparent how p00r of a substitute the attempt was.

  4. 1

    As Judge Newman’s dissent notes, a difference between an IPR-petitioned claim being denied an IPR trial and a claim that is taken into an IPR trial but not rejected is the absence of estoppel for that claim for other patent or publication prior art.
    But in both cases does it not seem highly unlikely that such a claim would be held invalid in D.C. litigation, when even the PTAB would not do so? So as a practical matter how much difference will it make?

    1. 1.1

      PM: a difference between an IPR-petitioned claim being denied an IPR trial and a claim that is taken into an IPR trial but not rejected is the absence of estoppel for that claim for other patent or publication prior art.

      Just to tease this out a bit more, let’s say my petition says “Claim 1 is obvious in view of W, X, and Y. Dependent claim 2 is obvious in view of W, X, Y and Z (Z addresses the additional limitation in claim 2).”

      PTAB accepts the petition with respect to claim 1, but not claim 2. Ultimately, I end up losing the argument. PTAB deems claim 1 to be valid over the prior art.

      I would be estopped in district court from arguing that those references render claim 2 obvious … right?

      1. 1.1.1


        Since claim 2 was not accepted into the IPR process, there is no estoppel on that claim.

        You do not get to parse the claim into its “items from claim 1” and “differentiating items particular to the dependent claim” and de facto rule that “pieces of” claim 2 are obvious.

        This is not to say that those references in addition to any other additional necessarily references cannot be used to make the argument that the claim is obvious, but there is no defacto finding based on what happened in the IPR for the accepted independent claim.


          You do not get to parse the claim into its “items from claim 1” and “differentiating items particular to the dependent claim” and de facto rule that “pieces of” claim 2 are obvious.

          I have no idea what you’re trying to say here.

          In my hypothetical, the PTAB has ruled that the broader claim is valid in view of W, X and Y. Given that, how can it possibly be the case that a narrower dependent claim with one additional element is obvious in view of those same references plus one reference directed only at the additional element? Why on earth shouldn’t estoppel apply here (taking the general estoppel effect of the statute at face value)?


            It can happen because that dependent claim is evaluated on its own and as a claim ad a whole.

            Each claim is viewed on its own.

            Thus, no estoppel.

            (This is in about as short and declarative English sentences as you can get)


              that dependent claim is evaluated on its own and as a claim ad a whole.

              Can you give an example of an instance where an independent claim (“X + Y”) is non-obvious in view of reference A but a narrower claim depending from that independent claim (“X + Y + Z”) is obvious in view of the same reference? Assuming the answer is no, what difference would it possibly make to the validity of the dependent claim if there is an additional reference cited only for the proposition that Z is an obvious addition to X+Y?


                You want to arrive at an Ends while the discussion is about the Means.

                Slow down and pay attention to the discussion of the Means.

                1. the discussion is about the Means.

                  Nice try.

                  The discussion is about basic logic, which you seem to have great difficulty with.

                2. There is NO “nice try” as my position fully comports with basic logic.

                  You are doing your number one meme thing again of:


          Yes, this is where Ned went astray in a blog below in arguing for no IPR examination for amended claims where the amendment is “merely” adding a dependent claim into another claim in the IPR. There is no legal difference between a dependent claim and an independent claim with the same combined limitations!
          Furthermore, dependent claims are, of course, not necessarily just adding obvious additional features or species details, etc.
          In any case, as demonstrated, not handling dependent claims seriously enough in an IPR petition can get them refused IPR consideration, and some folks apparently consider that a very desirable PTO validity “whitewash” for dependent claims arguably infringed.


            Actually Paul, that was Ned’s low hanging fruit and the only thing that he was close to being correct upon (for example, in the instance that the independent claim is not brought in an IPR).

            As pointed out earlier, if the independent claim falls in an IPR, there really is examination needed to find the line along the spectrum of the now bounced independent and any previously allowed (due to that previous independent) dependent claim.

            This level of detail just was not thought through by Congress.


            Paul, you seem to skip over the requirement that claims added in an IPR cannot broaden in any respect. They have to narrow. They must have all the limitations of an existing independent claim, and then have something further added.

            I also made the point that in most cases, dependent claims are simply allowed in ex parte prosecution if their independent claims are allowed. In court, these “unexamined” claims are still presumed valid regardless that their parent claims are held invalid.

            There would be nothing new in allowing them in an IPR if petitioner does not object. Someone, another petitioner or a defendant in court, still has the burden of proving invalidity.



              You are the one here who has ig nored the counter points presented to you on the topics you raise yet again.

              Shall I point out your errors again, only to see you go quiet?


              Ned, of course I knew that, but you are ignoring the fact that when any amended claims are introduced into an IPR they must also show full 112 support, priority date benefit, etc. That may well differ from the parent claim.
              Furthermore, your valid point that dependent claims are frequently not even separately examined in the original application prosecution is actually demonstrating another good reason why such claims, which were not even addressed in the IPR petition, should be examined in an IPR before being re-allowed with a PTAB decision blessing and given estoppel benefit.


                Paul, I have never once ignored that the Patent Owner needed to show support. Never once.

      2. 1.1.2

        As understood, there is no estoppel for a claim for which no IPR was actually declared, even though requested. [Wasn’t there a recent Fed. Cir. case on that?] Which I understood is what J. Newman’s dissent was complaining about. She is assuming the defendant would be free to make the same prior art arguments against that same claim in the D.C. litigation [because of no estoppel] [and arguing that this is inefficient]. I am merely suggesting that that would likely be a futile effort anyway. I.e., if one cannot make a good enough invalidity pitch to convince PTAB APJs to even take your case for that claim, how likely is an invalidity jury verdict for you on that claim?


          I do not see why you observation about the futility of the argument is relevant to Judge Newman’s point. Congress intended IPR to be something that you do instead of district court litigation, to make the whole process faster and less expensive. If IPR instead becomes something that you do in addition to litigation, then far from making the process faster and less expensive, one actually makes the entire process (at least slightly) slower and more expensive. That would be perverse, and statutes are supposed to be interpreted in such a way as to give effect to the good that Congress intended to advance. New York State Dept. of Social Servs. v. Dublino, 413 U. S. 405, 419-420 (1973) (“We cannot interpret federal statutes to negate their own stated purposes.”)

          I think that Judge Newman gets this exactly correct. This is not the most important problem floating around in the patent system right now, but when presented with a chance to make the IPR system work more well or less well, one would like to think that the CAFC would choose “more well.” Unfortunately, it seems that this is only true of Judge Newman.


            Yes Greg, and my point was simply that as to those claims which the PTAB says the petitioner does not even present an adequate invalidity case for an IPR, is there real duplication? There is no IPR trial on those claims, and are defendants really going to be able to successfully defend D.C. litigation on claims for which the PTAB says they did not even present a viable invalidity case? Unless they come up with new and better prior art for the D.C. are they even risking S.J. and/or sanctions?

            Also, what if a Petitioner has simply failed to make any effective argument at all for the invalidity of some of the claims it is listing as challenging? How can the PTAB legally institute an IPR on those claims?

            [I am not arguing the statutory language issue here.]


              This also gets back to the fact that the PTAB is not the Examining Corps. In an IPR that is expressly the IPR Petitioners job. The IPR statute makes that clear:
              “(e) Evidentiary Standards- In an inter partes review instituted under this chapter, the petitioner shall have the burden of proving a proposition of unpatentability by a preponderance of the evidence.”


                The “expressly the IPR petitioner’s job” statement is not accurate in the context here.

                As has been pointed out, the only true examination that can lead to grant is examination by the Office (Tafas remains good law on the point that the Office cannot force examination to be done by any third party).

                1. Anon, if that were actually true for an IPR, the PTAB would have to refuse entry into a patent of ANY amended claims in an IPR other than for technical corrections.

                2. Well Paul, that too would be a difficult point for allowing IPRs to continue.

                  The one thing that Ned had stated previously which is contradicted by what he says here is in regards to dependent claims.

                  ALL claims are supposed to be fully examined prior to grant. Yes, this includes narrower dependent claims.

                  For Ned to hide behind an error as if that error excused the error under discussion here is problematic at best.


              Right, if they failed before the PTAB then they will probably fail before the D. Ct. The mere fact that they are even allowed to try in front of the D.Ct. defeats Congress’ purpose in creating IPRs. It is no answer to say “let them knock themselves out, they will fail in the end.”


                Hm, as I re-read this, I see that I left out a “however” that would probably have made the post more clear. I meant to say “The mere fact that they are even allowed to try in front of the D.Ct., however, defeats Congress’ purpose in creating IPRs.”

                That is to say, the fact that they will like lose in front of the D.Ct. is really no answer to Judge Newman’s point. You really want all 102/103 issues to be decided at once, so that the process is made more efficient. Instead, the CAFC has made a dog’s breakfast of the system, by reading the non-reviewability provisions expansively, but the estoppel provisions narrowly, so as to assure that IPRs result in more, not less time and expense in litigation. This is a perverse outcome, but regrettably Judge Newman appears to be the only active service judge who sees this.


            …and yet, the total Article III (albeit not just the district courts) has indeed skyrocketed since the AIA – as noted several times now.

            Something else besides the takings infirmity when considering IPRs…


              Not quite true, since new D.C. suits are down considerably this year, new IPR petitions have leveled off, and the number of IPR Petitions denied has increased. Approximately 80% of IPRs are in direct response to D.C. patent suits. The thousands of claims killed in IPRs cannot continue to be asserted in those D.C. patent suits. How can that not be a very significant patent litigation cost savings for all those defendants?


                Not sure that that was what Prof. Crouch’s graph (sometime around the end of August?) actually showed Paul.


              Exactly. Judge Newman is a national treasure, the most natural heir to Judge Rich, and the judge in active service still who best understands why we even have a patent system.

Comments are closed.