The New Role for Post Grant Review Proceedings (PGR)

by Dennis Crouch

It has now been just over 18-months since we started filing patent applications under the new first-to-file patent system mandated by the America Invents Act (AIA) of 2011.  And, slowly but surely, FTF patents are now beginning to issue.  Those new patents also bring us into a new era of Post-Grant-Review-Proceedings (PGR).

My expectation is that PGR will quickly prove itself as an incredibly powerful tool for challenging patents. Importantly, PGR proceedings allow a third party to challenge the validity of a patent on any of the validity grounds that could be raised in court as a defense under 35 U.S.C. 282 – these include novelty, obviousness, written description, enablement, indefiniteness, utility, and subject matter eligibility. This list represents a major expansion of challenge-justifications over inter partes review (IPR) proceedings or reexaminations. These same doctrines have already been available for covered-business-method review (CBM) proceedings and have been used successfully if only on a narrow class of patents (those covering financial business methods).

Two recent PGR proceedings have been filed.  In one case, LaRose Indus. has challenged Patent No. 8,684,420 that covers the popular children’s “Rainbow Loom.”  Although the patent claims a 2010 priority date, the particular application was filed as a continuation in July 2013 and, at least according to the petitioner, added new mater to the claims at that point.  The second case PGR case involves Patent No. 8,598,291 that covers a liquid formulation of a particular nausea control drug used by chemotherapy patients. The petition argues that the claims fail the requirements of written description, enablement, and definiteness.

The most important limitation on the PGR process is the relatively short 9-month window for filing a PGR petition. Namely, the petition must be filed within 9-months of the patent issuance.  This puts some pressure on parties to monitor competitor patents and patent families.  It also provides patent owners at least some incentive to hold-off asserting their patent until after that window has closed (unless the patentee would welcome the review).  I suspect that the set-up in the LaRose Rainbow Loom market will be fairly common. The patentee (Choon) had previously sued LaRose on a previous patent and so LaRose has been closely monitoring the continuation applications.

= = = = =

In thinking about how soon the AIA cases will come on-board, I created the chart below. The chart provides some information on the timing of patent issuance — showing the percentage of patents issued with 18-months of filing and 36-months of filing respectively.  A big caveat for this particular chart is that it looks at the actual application filing date and does not consider the priority date that will typically control whether a patent will be examined under the AIA. Thus, even at the three-year date (March 2015), I would expect that significantly less than half of the recently issued patents would be FTF patents.

You’ll note the trend beginning in FY2011 of issuing more patents more quickly.  I largely attribute the change to two initiatives pushed by David Kappos during his term as PTO Director that resulted in (1) more funding to hire more examiners; and (2) a greater allowance rate.  Those factors have continued through the succeeding interim leadership terms of Terry Rea, Peggy Focarino, and now  Michelle Lee.


59 thoughts on “The New Role for Post Grant Review Proceedings (PGR)

  1. 6

    You’ll note the trend beginning in FY2011 of issuing more patents more quickly. I largely attribute the change to two initiatives…

    My friend Occam just dropped by and added another initiative to attribute the trend to: the large scale switch to address new apps (you know, to meet that first guarantee of 14th month action on the merits) and the re-configuring of putting RCE’s on the back burner.

    As somewhat recently captured here – this type of moving the deck chairs has itself changed again – so I would expect that the trend does not in fact continue (no matter what leadership team, properly vetted by the Senate or not, is in place).

    1. 6.1

      Thanks Anon – I agree that the USPTO has been playing games with its dockets. First shifting away from quickly considering RCEs and now switching back toward looking at them more quickly (after seeing the RCE backlog grow so rapidly).

      However, I do not think that particular ‘deckchair shifting’ had much of any positive impact on the number of allowed cases. In fact, if you want to quickly get more cases allowed, RCEs may be a good focus because many are ready for allowance by that point.

      1. 6.1.1

        If you really want to see a graph with an impact, run the allowance rate graph, which has now returned to its ‘traditional’ 70% allowance – the graph that so clearly captures the “Just-say-No” attempt by the Office to Reject-Reject-Reject its way out of the building backlog a few years back.

        Then, if you really want to see a graph with impact, flip that graph of Reject=Quality (as was boasted by Office personnel) and introduce an appropriate time lag and compare to the mounting backlogs in the RCE and Appeals queues.

        Yes, professor, the deck chairs analogy is spot on.


          Do you have a link to said graphs? Or a data set? I agree it would be interesting. Was a massive delay in allowances the only thing accomplished by the push for “quality?”

  2. 5

    Anyone know whether the 9-month cut-off date is “iron clad” unconditional; such that any new and / or amended claims submitted / entered after the 9 month cut-off date could not be subjected to a PGR?

    1. 5.1

      If you do a re-issue, then the 9-month is restarted.

      35 USC 321(c) Filing Deadline- A petition for a post-grant review may only be filed not later than the date that is 9 months after the date of the grant of the patent or of the issuance of a reissue patent (as the case may be).

  3. 4

    I would take bets that the use of PGR will be miniscule as compared to IPRs, [other than perhaps some generic drug maker patent challenges]. First, because only a small percentage of patents are sued on within only 9 months of their issue date. Companies simply do not go arouind instituting expensive and very dangerous contested case proceedings like this if they are not being seriously threatened by a patent. Very dangerous because the draconian PGR estoppel provision will essentially leave the petitioner defenseless to raise almost any future validity challenge. [If they have good patent or publication prior art they might be far better off saving it to use a few months later in an IPR, without distroying other possible defenses.]

    1. 4.1

      One comparison point is the EPO where about 5 to 6 % of patents are opposed by a third party. That system also has a nine-month deadline for filing a notice of opposition. This suggests that many companies have developed mechanisms for identifying potentially problematic patents held by their competitors. The US PGR rate may be lower based upon (1) the higher cost as compared to Europe; (2) the ongoing availability of alternative mechanisms in the US for challenging the patent once things get serious that sometimes don’t work well in Europe; and (3) the fact that a higher percentage of US patents are obtained by small or no-name entities (making them harder to track). However, the patent game is often higher-stakes in the US – which would tend to raise the rate here.

      1. 4.1.1

        Unless there has been a drastic change since my last one, EPO oppositions are nothing at all like the interference-trial-like evidenciary TRIALS with conflicting declarations, live witness cross-examinations and strict judicial-like rules of USPTO IPRs and PGR’s. You submit an EPO opposition paper with prior art patents or publications, and then, eventually, you get a hearing. Not much more to it normally. The patent application owner can easy submit amended or even alternative/contingent claims.


          Further demonstrating that EPO oppositions are not at all validly comparable to U.S. PGRs or IPRs [a point I have tried to make for years with U.S. PTO officials making invalid comparisons], this is an actual quote for an average in-house UK EPO opposition cost estimate I obtained in 2011: “£7500 for UK EPO patent agent time. Allowing travel and accommodation costs of say £500 and the EPO opposition fee of 705 Euros (£620), we get to a total of £8620.” [Versus $250,000 often quoted for a current IPR!]


            Furthermore, given the huge number of issuing patent claims these days, many if not most companies, especially in electronics and software products, no longer consider it even feasable, much less cost effective, to even do “watches” of all issued patents that might possibly impact some present product much less some possible future unknow product.


              Further, the EPO grants a Certificate registrable at the national Patent Offices of 40 European States, and opposition at the EPO your last chance of a central attack to kill it in all 40 States. Miss that window and you have to assaault validity separately in each EPC Member State where the patent is a problem.

              And as you say, it is cheap.

              And still only 5% get opposed.

    2. 4.2

      Paul, I predict that PGRs will primarily be filed against continuations and reissues of enforced patents.

  4. 3

    Thinking about Larose’s challenge, wouldn’t the patent office treating a con application on a pre-AIA basis (as I believe they legally must) preclude a third party from treating the con on a post-AIA basis?

    Any difference if instead of a con, it was a CIP?

    An IPR/CBM on the “C” ; a PGR on the “IP?”

    1. 3.1

      The argument here (made in the petition) is that there was actually new matter added to the claim in the CON – making it actually a CIP.

      During prosecution, the PTO treated the case as a continuation and, as such, examined the case under the pre-AIA rules (b/c of the earlier priority date). It will be interesting what weight the PTAB gives to that prior examination determination when determining whether this is actually an AIA case.

      1. 3.1.1

        So if Larose prevails in having the con treated as a cip (a very high bar in my estimation) … has the door been blown open to PGR-attack all the claims of this patent … even those claims which contain no new matter?

        A new risk for cips … and “arguable” cips?

        Going forward, seems best to file all-new apps instead of cips; incorporating by reference the / any con(s) in the new apps.


          Right – the test is whether the application ever contained a claim with a post-AIA effective filing date. If so, then the entire patent is treated as an AIA patent.


            Thanks. Does that mean that any further cons are treated as AIA patents as well … even if it/they do not contain a claim with a post-AIA effective filing date (if that’s even possible)?

  5. 2

    I take it that the petitioner for PGR has only to assemble a preponderance of evidence in order to succeed.

    I predict that disputes will become attritional. Each Petition for PGR will result in another patent family filing at the USPTO, that takes into account all the validity attacks of the PGR. When that new family member issues, it will in turn attact another PGR.

    1. 2.1

      I think that is probably true. I also think that PGRs are going to be rather limited. Probably only a few areas will bother to do this such as pharmaceuticals. Another area will be patents issuing off of continuations off of litigated patents.

      My prediction—very low usage. 10% at best of IPRs.

      1. 2.1.1

        NWPA – regarding you prediction of 10% of IPRs: In your view, is the limiting factor simply the 9-month window? I agree with you that it is pretty tight.


          Not just the 9 month window (which is a huge factor), but the speculative nature of it. Probably no one will assert the patent until the 9 month window is past. So, a PGR will have to be speculative, and I don’t see many companies doing that. Some companies will. And there will be special cases like continuations from licensed or litigated patents, but I think this is going to be small.


            Robin Feldman did a study on recent lawsuits and found that >25% of infringement complaints were field within two-years of the patent issuance with most of those coming within the first year following issuance.


              I suspect almost all attorneys are going to counsel a client not to assert a patent until the PGR is over.


                Like in Europe then. But note that more than 50% of all oppositions at the EPO are filed by Germans. Why? Because of Germany’s “bifurcated” patent litigation system, in which you have to fear being enjoined by a patent that, years later, gets declared to be invalid.

                So will Americans behave like shrewd German companies, and get their retaliation in first?

                Note that in Germany the presumption of validity in the Courts is a preponderance, not a C+C standard. That fact might suggest that there ought to be more PGR than German-originating opposition activity at the EPO.

                1. But they can still use IPR or CBM if the patent is enforced after the PGR period.

    2. 2.2

      Max, that can happen, and it is a known litigation tactic, but outside of pharmaceutical companies there are not that many companies willing to pay for such plural sequential continuation games for plural patents that would attract that many seriously threatened parties willing to pay for and risk the PGR estoppels. Also, the longer the application pendancy due to plural continuations from the earliest claimed filing date, the shorter the patent term. Plus there is a seriously enhanced danger of claim invalidity for “obviouness type double patenting” between the different “family” applications.

  6. 1

    While reduced pendency is good, it can’t be sought in a vacuum. Often the best prior art is published around the same time as the application is filed, so delaying prosecution within reason allows better search results, and theoretically ultimately better patents.

    That said, the current balance still favors a policy of reducing pendency. I think about 20 months to first action (from priority date, not US filing date) is about right, but that’s just my personal opinion.

    1. 1.1

      I agree that our current system works best if the examiner does a search that is at least 18 months from the application priority date to capture the prior-art in previously-unpublished application filings.

      1. 1.1.1

        Yes, by all means, lets be sure to look at the secret, unpublished at the time of filing, “prior art”.

        By the way, Re: the suggested 20 month FOA. As I recall, at one point someone (maybe Dudas) declared that it was the goal of the PTO to issue a First Office Action on the Merits, within 14 months of filing. I believe that is still the “goal”.


          A couple of points:

          1) One reason why you might expect a higher allowance rate for track-1 (accelerated) cases is that some of the prior art is not available (formerly 102(e) prior art, now 102(b)(2) prior art).

          2) I have a few reasons for thinking that the 14-month FAOM (First-Action-on-the-Merits) is a good goal. First, getting patents out more quickly is helpful for the system. Second, since most applications (~70%) claim priority to a prior application, the 14-month goal is usually well past the 18-month-from-priority publication date. Finally, the fast majority of applications require a second search that will be a few months down the line.


          By the way, Re: the suggested 20 month FOA. As I recall, at one point someone (maybe Dudas) declared that it was the goal of the PTO to issue a First Office Action on the Merits, within 14 months of filing. I believe that is still the “goal”.

          Huh. I wonder if he was aware of the fact that Congress had already specified this 14-month “goal” in 35 USC 154.


            Hehehe. Maybe the fact that it is in my experience almost never met, by a factor of about 2, lead me to forget that it was statutory.

            Thanks for the reminder.


          Yes, by all means, lets be sure to look at the secret, unpublished at the time of filing, “prior art”.

          Well the flip side of that logic must be that if someone else creates an embodiment before the publication of the application, that must prove it is an obvious change. One of ordinary skill made the invention without the assistance of the published application, right?

          Nah, pretty sure both that statement and yours are wrong, as the standard is first to file, and “prior art” unpublished applications at the time of invention were still filed first.


            “Well the flip side of that logic must be that if someone else creates an embodiment before the publication of the application, that must prove it is an obvious change. One of ordinary skill made the invention without the assistance of the published application, right?”

            So I invent something. I file an application on it. Before my application publishes, somebody else “creates an embodiment” of my invention, then that “proves” my invention is obvious because somebody else created an embodiment without the assistance of a publication of my application?



              In an article, John Duffy suggested that independent invention that occurred immediately following your invention should logically be seen as a factor in the obviousness determination for your invention. However, I believe that Duffy would agree under the current statute that evidence should not be considered as prior art.


                Thanks. I’ll look it up and read if it’s available. Lemley also wrote an article on “simultaneous invention” (or whatever he called it) that I read and found unconvincing. As Les correctly notes below, section 103 is there to prevent patents on inventions that would have been obvious to one of ordinary skill in the art, not inventions that are the creation of those of extra-ordinary skill in the art.


                Prof. Duffy is only human and this is a clear miss.

                A race to the patent office actually (shocking as it sounds) may include more than one racer.

                The de facto notion that more than one racer implies that NO ONE wins the race is most bizarre. If Duffy were correct, then even under the old laws, we would have no need for interference practice.

                1. As far as I know, with Duffy’s “Timing” theory of obviousness is that one more factor to consider is how long it took the herd to latch on.

                  Suppose there are several independent filings, all around the same time, for an improved snowboard, improved in that a particular proprietary DuPont polymer is used as a board coating. Does it make any difference to obviousness if the DuPont market launch of this particular polymer took place just 3 months before the cases were filed, and the market launch emphasized the slippery but hard-wearing properties of the new polymer grade.

                  Would it be any different if the DuPont market launch was 10 years before Burton Snowboards filed, and so far only Burton has filed. Duffy thinks it would. What’s so unreasonable about that?

                2. Your theory misses MaxDrei, as this is expressly NOT a herd latching on type of situation – these are contemporaneous racers – not one racer out in front with a herd seeing that and latching on.


            1) Which “statement” of mine is allegedly wrong?

            2) Filed first doesn’t mean invented first. All we know is it was filed faster.

            3) Why do you assume that the second inventor is only of ordinary skill?

            I say on a planet of several billion people it might be that two or ten people of extraordinary inventive brilliance might turn their attention toward the same or similar problems at a similar point in time and separately arrive at similar or the same solution. That does not mean the solution was obvious to those of merely ordinary skill in the art.


              You do not even need to go there Les.

              The last sentence of 103 was added for exactly that reason – there is no need to even reference “extraordinary inventive brilliance” (or even a Flash of Genius).

      2. 1.1.2

        The 18 month window is outdated and should be shortened. I can’t think of a single software claim that wasn’t in fact created by others within the 18 month window (and thus the public did not in fact need the disclosure, which makes it an unnecessary monopoly).


          So change the 18 month publication for all technologies because for one technology it may be “outdated.”

          Sounds like a great proposal.

          “I can’t think of a single software claim that wasn’t in fact created by others within the 18 month window (and thus the public did not in fact need the disclosure, which makes it an unnecessary monopoly).”

          Is your personal recollection all that we have to go on here, or do you have something more objective that we could review in evaluating your proposal to shorten the 18 month publication window?


          The guy who is clueless as to why the term is even there has no ground to say that the length of the term is “outdated.”

          Clearly, given the prodigious amount of Patent Term Adjustment doled out by the Office for missing this guarantee, not only should the length of the term NOT be shortened, but likely – to preserve the reason why Congress put that term in there – it should be lengthened.

          Let’s use our critical thinking skills people.

      3. 1.1.3

        Maybe this is already the case, but it seems to me that the PTO ought to be able to search unpublished applications for art that will later be published and become prior art. In the relatively infrequent case where that’s the most relevant art, if they have to issue another non-final office action or even sit on the first action until the application publishes, so be it.


          it seems to me that the PTO ought to be able to search unpublished applications for art that will later be published and become prior art

          We cannot. As a matter of fact we can’t even search the application we’re examining before it is published, which makes examining non-published applications a real b*. It’s just swell to lose your first half hour of examining OCRing, then fixing the OCR mistakes, then reformatting the claims of your application just so you can begin to work with them for mapping. God help you if there’s a preliminary amendment because strikethru and underlining are too difficult for our OCR.


            I have never heard of a PTO being able to search amongst as yet unpublished applications. I expect never ever to hear of it.

            In Europe, Patent Offices vary, on whether they will issue without first doing a full search for everything filed one day before, but not published till 18 months after the date of the subject claim. The UK PTO will issue early, the EPO not.

            Take the case where the EPO is the ISA for a PCT filing, and searches imediately after the PCT is filed, to issue ist WO-ISA. Downstream, in the EPO national phase, the EPO will not issue till it has done what it calls its “top up” search, done to capture all that “18 month later” stuff that simply couldn’t be searched in the preparation of the WO-ISA.


              Yeah, the “top up” search makes sense, and maybe it’s prescribed in the USPTO, I don’t know. I suspect that even if it is, that it is not done with any rigor, since by that time the examiner simply wants to close the case, and certainly doesn’t want to have to issue another non-final OA.



            You used to be able to search the unpublished applications. The PTO had an internal database called UPAD that you could select in EAST just like the US-PGPUB, USPAT, and other databases. The PTO, in their infinite wisdom, chose to discontinue this database back in 2010 I believe.

            Unfortunately, imho, you are poorly served by an IT department that does not consider database strength to be a high priority. For example, at the beginning of the year they uploaded the FPRS database. This is an incredible treasure trove of information, but was implemented in such a ham handed manner that hardly anyone ever uses it.

            There are plenty of other prior art (patent) database choices that the PTO has made that baffle the mind, but none of the attorneys on here want to read a thread hijacking like that…


              The subject of those choices and how the PTO goes about making them would be of interest to many here.


                Thanks for asking. How the PTO goes about making those choices is a complete and utter mystery. I’ve talked to the decision makers and they explicitly tell me that zero input from experts on the outside is needed or wanted.

                1) They are building the new PE2E system using contractors that have never, ever worked with patent data. It’s taken Thomson, Lexis, STN, Minesoft, and Questel decades to get their systems to the level that people will pay for them. Hundreds of companies have failed over the past two decades trying to get into the patent analysis, data, and search space.

                2) They put a bizarre database of Chinese machine translations on the USPTO website and pretend that examiners use it. No one uses it. You can’t use it to do anything and it doesn’t integrate with a single tool in the examiner’s tool belt.

                3) If an examiner wants to search the first publication of many US filings, they need to go to PatentScope (WIPO) or Espacenet (EPO) on the web to do so. The USPTO DOES NOT HAVE RECENT PCT DATA IN THEIR DATABASES! Yes, I’m yelling mad about that.

                4) From 2010 to 2013 they completely stopped uploading European data into their computer systems. Zero data. Not even the Abstracts.

                5) At the beginning of this year, they finally uploaded all the European data into their FPRS database. Unfortunately, they took the nice, convenient family grouping the EPO uses and loaded everything as individual records. But only one of the family of records actually has data. So when you search a subclass, you might get six blank records for every record with an actual document. And you can’t sort the data to eliminate this. After clicking through blank pages for twenty minutes, 99.99% of humans deactivate this database with zero regrets.

                6) Don’t ask me how much full text the examiners can search. Most commercial search engines, like those by the companies listed in #1 above, have approximately twenty countries covered. Additional countries are covered by full text machine translations which can be turned on or off depending on how much noise you want to review. The US? There is a reason examiners generally cite only US art.

                Okay, that’s all I’ll do off the top of my head. I can keep going, but it’s bed time…


            As a matter of fact we can’t even search the application we’re examining before it is published

            This cannot be true, as this would mean that the 14th month guarantee of first action on the merits would NEVER be able to be met.

            You need to understand your job just a little better. May I suggest that you do so before preaching?



              He means that the unpublished case he is working upon is not findable in the databases to which he has access.

              One of the negatives of this is the lack of positive reinforcement provided by finding the subject application (or patent when doing a validity) while executing a search query. Is this a relevant subclass… the subject document is in there. Is this a good text based search… well, it found the subject document. Is this a good Class and key word combo search… at the very least, it found the subject document. Searching is very iterative process and this positive reinforcement helps the examiner or searcher know they might be on a good path.


              Of course it can be true. If the database only has published documents in it, then he can’t search the unpublished document in front of him using that database can he?

              He has said that in order to search an unpublished application he has to use an optical character recognition tool (apparently to convert the pdf file he has to searchable text) and then do his search. However, since the OCR software is not the best, he also has to correct the errors in the OCRing if he wants reliable results. This is not practical.

              That boils down to: he cant search the application in front of him.


              Anon, I believe what RandomGuy is saying is that a text version of the application is not available to the the examiner’s at that point. Many examiners like to use a text version because they can simply cut-and-paste to search for the words actually used in the claims rather than retyping.

      4. 1.1.4

        …works best?

        For whom?

        This is pure malaraky. What any objective view would keep in focus was that the ability to publish prior to grant – at all – was a degradation of the original Quid Pro Quo exchange.

        The 14 month written guarantee sought to expressly preserve the option to the applicant for expressly abandoning their application based on a full and complete first action on the merits. This was expressly prior to the 18th month mark in order to preserve the Quid Pro Quo deal – no free Quo (by publishing) without the Quid. With express abandonment preserved as an option – prior to publishing – the applicant can safely test the patent waters and preserve any trade secret aspects.

        Beware the bias of taking without giving.

Comments are closed.