Amgen v. Coherus: Argument-based prosecution history estoppel

By Jason Rantanen

Amgen Inc v. Coherus Biosciences Inc. (Fed. Cir. 2019)  18-1993.Opinion.7-29-2019

Judges Reyna, Hughes, Stoll (author)

The doctrine of equivalents (“DOE”) may be every patent law student’s favorite doctrine to hate.  (Edit: to clarify, I mean that it can be complicated not that it’s conceptually a “bad” doctrine.) I sometimes introduce it as “the rule against perpetuities – except more complicated and economically significant.”  Under the DOE, a patent can be infringed even when the accused product or process doesn’t literally meet all the limitations of the claim.

Central to the DOE is prosecution history estoppel: a doctrine that prevents “a patentee from using the doctrine of equivalents to recapture subject matter surrendered from the literal scope of a claim during prosecution.”  Slip Op. at 8, quoting Trading Techs. Int’l, Inc. v. Open E Cry, LLC, 728 F.3d 1309, 1322 (Fed. Cir. 2013).  Because prosecution history estoppel is a question of law, it often functions as threshold question that is resolved before moving to the fact-specific equivalents analysis.  Indeed, David Schwartz, Lee Petherbridge and others have argued that PHE is one of the key mechanisms that the Federal Circuit has used to limit the application of the doctrine of equivalents following Warner-Jenkinson.

Often when we think about prosecution-history estoppel, we think about amendment-based estoppel, which can arise when an applicant makes a narrowing amendment to a pending claim.  This case involves one of the trickier aspects of PHE: argument-based estoppel.


Amgen sued Coherus for infringement of Patent No. 8,273,707, which relates to methods of purifying proteins using hydrophobic interaction chromatography (“HIC”).”  Slip Op. at 2.  This process uses a combination of a buffered salt solution containing the protein that is poured into a HIC column.  The proteins bind to column matrix and impurities are poured off.  The ‘707 patent claims a process that increases the maximum amount of protein in solution that can be loaded into the HIC column by using certain combinations of salts “chosen form one of three pairs: citrate and sulfate, citrate and acetate, or sulfate and acetate.”  Slip Op. at 4.

During prosecution (this gives rise to the PHE ruling), the examiner rejected the claims as obvious in view of a prior art patent that “disclosed several salts for improving hydrophobic interactions between a protein and the column matrix.” Slip Op. at 4.   Amgen responded by pointing out that “the pending claims recite a particular combination of salts.  No combinations of salts [are] taught nor suggested int eh Holtz et al. patent, nor [are] the particular combinations of salts recited in teh pending claims taught nor suggested in this reference.”  Id. at 4-5.  A supporting inventor declaration pointed out the three pairs above “reduced purification costs on a commercial scale as compared to using only a single salt.”  Id. at 5.  After another rejection, Amgen further pointed out that “choosing a working salt combinationw as a ‘lengthy development path’ and that ‘merely adding a second salt’ would not result in the invention” and the examiner allowed the claims.  Id. at 6.

In 2016, Coherus filed an abbreviated Biological License Application for an Amgen product.  Because Coherus’s protein purification method uses a chromatography buffer containing a salt combination, but not one of the specific combinations recited in Amgen’s claim, when Amgen sued for infringement in 2017, Amgen’s infringement theory rested on the doctrine of equivalents rather than literal infringement.  The district court subsequently granted Coherus’s motion to dismiss on the pleadings for failure to state a claim.

Federal Circuit: Argument-based prosecution history estoppel applies here

On appeal, the Federal Circuit affirmed the application of argument-based estoppel. “To invoke argument-based estoppel, ‘the prosecution history must evince a clear and unmistakable surrender of subject matter.'” Slip Op. at 9, quoting Conoco, Inc. v. Energy & Envtl. Int’l, L.C., 460 F.3d 1349, 1363 (Fed. Cir. 2006).  Here, the court held, argument-based estoppel applies “because Amgen clearly and unmistakably surrendered unclaimed salt combinations during prosecution.”  Id. at 9.  Amgen repeatedly pointed to the particular combinations of salts as important in distinguishing the prior art references.  As a result, “a competitor would reasonably believe that Amgen surrendered unclaimed salt combinations.”  Id. at 10 (internal quotation marks omitted).

Amgen’s primary argument, that it asserted other reasons for distinguishing the prior art reference, was not persuasive as “our precedent instructs that estoppel can attach to each argument.”  Id. at 11.  Nor was the court persuaded by Amgen’s argument that it’s last response before the claims were allowed did not contain the particular-salt-combination argument.  But “[t]here is no requirement that argument-based estoppel apply only to arguments made in the most recent submission before allowance.”  Id. at 11.  For those seeking to avoid this outcome, note the court’s final words on the subject: “We see nothing in Amgen’s final submission that disavows the clear and unmistakable surrender of unclaimed salt combinations made in Amgen’s January 6, 2011 response.”  (I’m curious: has any applicant ever disavowed an argument made earlier in prosecution?  This seems like a risky strategy for getting claims allowed.)


39 thoughts on “Amgen v. Coherus: Argument-based prosecution history estoppel

  1. 7

    You likened it to the Rule against Perpetuities. From a judicial standpoint, it seems more like the Rule in Shelley’s Case: the existence of which everyone seems to know, but the limits of which nobody really seems to understand.

  2. 6

    OT other than being very much pharma related: “The Trump administration said Wednesday it will create a pathway allowing Americans legally and safely to import lower-cost prescription drugs from Canada.”
    Even if one could still believe this sort of promise from this administration, what would be likely to happen? This would presumably mean infringements of unexpired U.S. patents on some major drugs, so what are the likely industry responses? Importer [not individual] patent suits, if reachable? ITC general exclusion order actions? Interesting issues.

    1. 6.1

      This would presumably mean infringements of unexpired U.S. patents on some major drugs, so what are the likely industry responses?

      Is it an “infringement”? Under the recent Impression Prods. v. Lexmark Intern., 137 S. Ct. 1523 (2017), if the patentee sold the product in another country, then it is not an act of patent infringement to import that particular lot of product into the U.S.

      Until recently, such products have still be excludable from import under FDA law. The Court made clear in Pom Wonderful LLC v. Coca-Cola Co., 134 S. Ct. 2228 (2014), however, that there is no private right of action to enforce the FDCA. In other words, if the U.S. government does not (as a matter of prosecutorial discretion) choose to enforce the FDA law exclusions of these imported drugs, I am not sure that there is much that the patentee can do to stop the importation.

      I suppose that one could lodge an APA suit on the grounds that the failure to stop these imports is arbitrary and capricious. I do not care to predict how that sort of a suit would shake out, however, given the fairly deferential standards applied in APA suits.

      1. 6.1.1

        Thanks dozens, Dozens, especially for the reminder of Impression Prods. v. Lexmark [overruling Jazz Photo].

      2. 6.1.2

        What do you think about this Bloomberg article on this subject today?
        “Under Canada’s drug licensing program, pharmaceutical companies agree to sell products at a fixed price and to guarantee supply. In exchange, Canada commits to not sell on the products it purchases across the border, said Duffin. And while those are federal rules, health and drugs fall under the purview of provincial governments who have “zero interest” in not having drugs available for their own citizens, she said.
        “Who is going to sell it to them?” she said. “I don’t think any hospital is going to start selling its drugs when it’s already scrambling to keep its supplies. Or is it going to be the same pharmaceutical companies that are selling them their drugs at an inflated price?“
        Furthermore, the price of generic drugs in Canada isn’t any cheaper than in the U.S. because generics don’t fall under the federal pricing system, added Adams. “We have some of the most expensive generic prices in the world.” ..

    2. 6.2

      [W]hat are the likely industry responses?

      I foresee two lines of response: short term; and longer term.

      In the short term, the logical response is to try to forestall or nullify this announcement. Certainly this administration has announced many policy bombshells that never actually materialize (e.g., Mexico tariffs) or that are briefly tried, but retracted a few weeks later (e.g., budget standoff shutdown). It is possible that the right efforts targeted at the right congressional and administration decision makers could make this go the way of the Mexico tariffs.

      In the longer term (if the announced proposal goes actually into effect), the next most sensible response would be to stop selling product in Canada. If there is no product in Canada, then the product cannot be exported from Canada to the U.S.

      If Canada were to authorize other entities than the U.S. patentee to manufacture the products for sale in Canada, then this would not trigger “first sale” exhaustion of the U.S. patent rights. This, in turn, would give the U.S. patentees a private right of action (patent infringement) that could be used to stop the importation, even if the Trump administration chooses not to enforce the FDA exclusions.

  3. 5

    Re the question at the end: “(I’m curious: has any applicant ever disavowed an argument made earlier in prosecution ..?

    Yes, rarely, but I have – in a pre-litigation situation where a divisional or continuation was pending and prosecution control was taken over by litigation counsel.

    1. 5.1

      So… you took over prosecution – was that after you relinquished your registration number?

      If so, you do understand the ethical considerations of practicing before the Office without proper authorization, right?

      Greg’s remark on the Hricik side of the blog comes to mind…


          I deconstruct your apparent dust-kicking along several facets in a post that somehow caught the filter.

          Bottom line though is that your accusation of me fabricating anything is plainly false.

          That you appear to choose to make false accusations on top of what appears directly from your own statements to be an un eth 1 cal unauthorized practice of law — and one YOU should know better, having chosen to withdraw your Patent Office registration, may well be a serious problem accompanying your choice to use the “borrowed authority” of posting in your real name.

          Nothing untoward here is of MY doing, as I merely reflect that YOUR statements might need more care.

    2. 5.2

      While I recognize that this thread is getting a bit old in the tooth, I was tracking down a different item and came across an old exchange directly on point to the “oopsie” by the “I use my real name” Paul F. Morgan.

      His post of July 30, 2019 had reverberated with me for some reason, and that is why I directly asked him for confirmation here (and rather unsurprisingly, received scorn instead of a direct reply.

      As it turns out, this is not the first time that our “real name” friend has EITHER made an apparent admission of impropriety, nor responded to HIS admission with scorn and disdain, as opposed to providing a simple and clarifying answer.

      I take you back to an exchange dating more than a year previous: circa April 2018 and the exact same “oopsie-accusation-run-away-from-clarification” from “real Name” Paul F. Morgan:

      link to

      Gutless wonders never cease. And apparently, appear to repeat themselves (inadvertently or otherwise).

  4. 4

    What’s “tricky” about this decision? This post seems pretty clear that the patentee said, “My patent doesn’t cover that,” and then sued someone for doing that. If accurate, that sort of behavior should be sanctionable.

  5. 3

    Maybe (doubtful, but just maybe), I am reading the phrase wrong, but:

    The doctrine of equivalents (“DOE”) may be every patent law student’s favorite doctrine to hate.” indicates to me that you are teaching this stuff wrong.

    The notion of DOE is to provide strength to patents, such that Infringers cannot weasel around claim terms based merely on the limitations of the human language. If this notion is something that you feel the students have some sort of inherent distaste for, then you more likely than not are indicating an underlying “patents are bad” vibe that the students are picking up on.

    Students interested in having a strong patent system would not bristle so at a notion that provides protection throughout the penumbra of protection for an innovation.

    1. 3.1

      My guess as to another reason [in addition to “notice” violation and increased infringement litigation ambiguity] for stating that “The doctrine of equivalents (“DOE”) may be every patent law student’s favorite doctrine to hate: ” It seems hard to reconcile the actual extensive substitution facts of its parent Sup. Ct. Graver Tank & Mfg. Co. v. Linde Air Prods., Inc. (1950), with the current Fed. Cir. DOE tests under the later decision of Warner-Jenkinson v. Hilton Davis Chemical Co., 520 U.S. 17 (1997). The Graver Tank “triple identity” test for equivalence [requiring that the accused product “performs substantially the same function in substantially the same way to obtain the same result”] is further limited by the “all elements” test for equivalence in Warner-Jenkinson v. Hilton Davis Chemical Co., 520 U.S. 17 (1997). Under the “all elements” rule, the doctrine of equivalents must be applied to each individual element of a claim, not to the invention as a whole. It is necessary to show that every element of the patented invention, or its substantial equivalent, is present in the accused product or process.

    2. 3.2

      Anon: I was referring to the difficulty of understanding the complexity of the doctrine, not its substantive desirability. Conceptually, the DOE makes a lot of sense as a recognition of the inherent limitations of language and the difficulty of drafting perfectly-worded claims. I explain the concept as being like a 1 vs. 100 game, with 100 people trying to design around your patent claim. The DOE helps to balance that out.

      The practical challenge of the doctrine, however, lies in the numerous legal constraints, from the various levels of PHE, to the all elements rule and element vitiation, to the dedication to the public domain rule, to the Wilson Sporting Goods “hypothetical patent claim” holding. Added to that is the relationship between § 112(6) equivalents and DOE equivalents which requires carefully understanding which term is being used when. All of these have good reasoning behind them, but taken together they become quite complex.

      1. 3.2.1

        Thanks Jason:

        Conceptually, the DOE makes a lot of sense as a recognition of the inherent limitations of language and the difficulty of drafting perfectly-worded claims.

        There – that was easy. What student would not love that?

        Ah, I see your point though: plenty of patent profanity has been built up around something beneficial to the patentee.

        Gee, we have seen that in other places as well, haven’t we?

      2. 3.2.2

        All of these have good reasoning behind them, but taken together they become quite complex.

        What you are addressing, really, is the incredible complexity of the US legal system itself, and the resultant effect of that complexity on how lawyers are trained.

        The limits on the doctrine of equivalents are based on “good reasoning” because, at the end of the day, the DOE is an equitable doctrine that attempts to correct for “unfairness” to certain patentees. Similar doctrines could be invoked in other contexts to account for that unfairness … and sometimes they are! And sometimes they are not.

        What’s actually “complex” is not the “good reasoning” underlying the limits on the doctrine (because these are the “good reasons” that are justifiable, and lead to outcomes that pretty much everybody agrees are “fair”). What’s “complex” is remembering all the case names, the particular language used by the courts, and the occasionally arbitrary content descriptors that are given to the various “reasons”. In a better judicial system, when an equitable principle is raised to limit some other equitable doctrine, it shouldn’t have to be the case that the party raising the principle has “case law” to support the principle. What should matter is whether the principle makes sense, and whether justice is reached by applying the principle. At that point, a clerk should be able to fill in the details about earlier cases if that is what is believed to make the world turn round.

        This sort of thing is, again, why the decision in Mayo v. Prometheus can never be overturned without completely destroying the system as we know it. You can’t overturn basic logic without overturning the credibility of the entire system.


          Your cognitive dissonance and focus on the “good” limits as if patent profanity were a “good” thing is noted.

          What a schlup.


              Yet again my shifty friend wants to provide a known false allegation as some type of “witty” retort, and should be well aware that the “wikipedia” line only can show his own ineptitude stemming from the Nobel discussion point, now quite a few threads back.

              My shifty friend is just not the brightest bulb in the lighting rack.


                Try reading more carefully, Green anon. I was not referring to when you cut and paste from a website other than Wikipedia. (A DIRECT CITATION! I’m still reeling! Who could have seen that coming?) Rather, I’m referring to your mistaken understanding of the DOE due to a mistake in Wikipedia, and you trying (unsuccessfully) not to own it. US patent law hard.

                1. Try reading more carefully yourself as our last exchange on your rather odd “gotcha” attempt with Wikipedia deconstructed your continuing saga and apparent desire for self-flagellation on this very topic.

                  …or did you slink away too quickly from that 270 or so comment thread?

  6. 2

    Is this why 90% of the “arguments” I see are simply “The prior art does not teach [claim + amendments]”?

    1. 2.2

      Is this why 90% of the “arguments” I see are simply “The prior art does not teach [claim + amendments]”?

      If the art doesn’t teach the amended claims, then there isn’t much point in saying more. But the amendment itself will result in some estoppel. If I amend “salt” to “a potassium salt” in order to get around the art, I’m not going to be able to go after anybody who uses some other salt (i.e., a salt besides a potassium salt) using the DOE. At least, that’s the way it’s supposed to work. The CAFC has written some jacked up opinions which might suggest otherwise if you read them with your eyes crossed but a good defense should be able to distinguish those.

      1. 2.2.1

        MM bruh did you catch the newest salvo in the fight against the evil of the white cis hetero christian capitalist patriarchial gubmit structure?

        link to

        I applaud her. Soon we will show the dems that diversity is a must, not a mere frivolity!


          Not clicking on that, whatever it is. But I did see the Repu-k-k-k-es doing whatever they can to encourage the hacking of future US elections. Golly, seems like a strange position to take but I suppose if you have no path to a viable majority, you might as well just destroy everything and take the spoils while the getting is good.

          Dennis and Jason get to see this right up close, of course. As do their employers. Thank goodness they are tenured and can speak truth to power. Oh but wait! “Campus PC” is suppressing them.



            Why are old timers afraid of NBC news? Sure it’s propaganda, but you’ve been “slurping it up” since you were 5 most likely.


            Here’s the headline for you: “Executive director of House Democrats’ campaign arm resigns; 5 others also leaving”

            Bottom line, the chief of dems HOUSE/congress campaign overall strategy up and quit because there wasn’t enough brown skin in the executive room. They are now said by themselves to be in chaos w/ november just a stone throw away.

    2. 2.3

      No. I expect this is due to super low budgets from clients, which have become common place.

      If it is clear that the prior art does not teach X, there is no use explaining why the prior art does not teach X. If it is not clear that the prior art does not teach X, then the applicant is half-asssing it and there is probably a low budget.

      There are ways to make substantive, persuasive arguments without shooting yourself in the foot later on.

      1. 2.3.1

        I think this is right. The budget continually has declined. I’d say it is about 1/3 of what it was 15 years ago.

        But also best practices teaches us to make an argument why it is not taught and then make arguments why it matters to the examiner on the phone. So win and leave no trace.


          Hmm, not liking the notion (and note that I am not disagreeing that such may happen) that “best practices” seeks to disregard 37 CFR 1.2.


            There are big corporations that I work for that would accept nothing but. SOP at this point. Also the top prosecution workshops teach this as best practices.


              Perhaps you misunderstand – I fully agree with conducting interviews, not for which at least one has the ability to interact with another human being by other than the written word.

              Such is most definitely part of “best practices.”

              But the view as to why that is a best practice is distinctly different if you are aiming to “win and leave no trace” as opposed to win and leave a solid trace as to WHY you won (quite apart and distinct from the advantage of dealing with someone on a more human and personal level).

      2. 2.3.2

        Maybe, but the fact that X is usually gibberish has nothing to do with budgets.

  7. 1

    Should not there logically be a difference between using prosecution history estoppel [PHE] from attorney application prosecution arguments to preclude the abnormal expansion of claim scope beyond any literal claim scope by DOE,* vis a vis using PHE to narrow literal claim scope, to avoid prior art, for which the Fed. Cir, has called for a clear disclaimer of record?

    *A;n exception to the public “notice” of infringement function of claims.

    1. 1.1

      Also interesting here is the inherent Fed. Cir. approval here of a granted D.C. “motion to dismiss on the pleadings for failure to state a claim” for non-infringement. I.e., presumably even earlier, faster and cheaper than getting a summary judgement decision of non-infringement.

      1. 1.2.1

        Yes, as noted, under the the public “notice” of infringement function of claims, which is only affected only by claim broadening beyond literal infringement, and further evidenced, for example, by the clear Congressional distinction between claim broadening and claim narrowing in reissues and the allowance in reexaminations and all other post grant proceedings of only claim narrowing.

Comments are closed.