Opening a Closed Markush Group

by Dennis Crouch

Amgen v. Amneal Pharma (Fed. Cir. 2020)

Amgen’s patent at issue covers a drug formulation for treatment of hyperparathyroidism in particular situations sold under the trade name Sensipar with about $1b annual sales. U.S. Patent 9,375,405.  Note here, that Amgen’s primary patent on the blockbuster drug expired in 2018, but this secondary formulation patent has been effective at keeping generics off the market.

Amgen sued Amneal, Piramal, and Zydus for infringement based upon their ANDA filings with the FDA.  The district court ruled that the Zydus proposal would infringe, but that the proposed formulations by Amneal and Piramal would not infringe.  On appeal, the Federal Circuit has vacated the verdict for Amneal on claim construction, but affirmed no-infringement by Piramal and infringement by Zydus.  Thus, Piramal has something of a green-light to move ahead.

A major portion of the appeal focused on proper construction of a Markush group that required a “binder”:

A pharmaceutical composition comprising: . . .

from about 1% to about 5% by weight of at least one binder selected from the group consisting of povidone, [HPMC], … , and mixtures thereof.

‘405 patent claim 1.

The Markush Group claim form stems from a 1923 decision by the Commissioner of Patents that allowed a patent applicant to effectively define and claim a genus by enumeration of species within the genus.  Ex parte Markush, 1925 CD 126 (Com. Pat. 1924).  In Markush, the patent applicant attempted to claim a diazotized compound for making dyes that includes

material selected from the group consisting of aniline, homologues of aniline and halogen substitutes of aniline.

The applicant (Markush) had considered a broader descriptive genus such as “mono-amines” but concluded that to be too broad since it including too many non-working embodiments.  The examiner though refused to examine the “alternative” claim form.  

In his decision, the Commissioner in Markush concluded that there are “many instances in which a generic term covering a number of substances cannot properly be employed” and that a listing of alternatives may be the only way the applicant “may cover his real invention without filing a number of applications.”

Note here that Amgen’s Markush group is remarkably similar to that presented by Markush’s patent attorney Victor Borst back in 1925.  The claim format has remained remarkably “stuck” as a limited mechanism for avoiding claim-in-the-alternative objections.

Amneal’s binder includes HPMC as recited in the claims, but also includes polyethylene glycol (PEG).  The major question in the case here is whether Amgen’s Markush Group is “open” or “closed.”  In other words, does the presence of the additional compound in the binder (PEG) avoid infringement.  The district court found the claim closed to additional binding agents and thus that Amneal does not infringe.  On appeal, however, the Federal Circuit rejected the claim construction:

There is no language in Amgen’s claim indicating that every binder … in the claimed formulation must be within the Markush group[]. Instead, the claim recites “at least one” binder … “selected from the group consisting of” various excipients. And the limitations merely require that those particular binders or disintegrants meet the specified weight-percentage requirements, which is not inconsistent with the overall composition containing other binders or disintegrants. The plain language of this claim requires “at least one” of the Markush members and certainly does not indicate that the only binders and disintegrants in the claimed formulation are those listed in the groups. And we do not see a sufficient basis for a different conclusion in the specification or in the prosecution history we have recited. . . .

[The claim] is satisfied when an accused product contains a component that is from the Markush group and that meets the limitation’s requirements for the component. It does not forbid infringement of the claim if an additional component is present functionally similar to the component identified in the Markush group limitation, unless there is a further basis in the claim language or other intrinsic evidence for precluding the presence of such additional components. There is no such basis here.

Slip Op. Thus, while the markush group itself may be closed (based upon its “consisting of” language), the claim as a whole is open (based upon its “comprising” transition in the claim’s introduction).  With this broader claim interpretation, it looks like Amneal will be found to infringe. 

Piramal’s proposed product is different in that its binder is a starch not listed in the Markush group.  Thus, no literal infringement.  Amgen unsuccessfully argued for infringement under the Doctrine of Equivalents.  The problem was that the Markush group was added as part of a narrowing Examiner’s amendment that led to an allowance. (The group was already in a dependent claim — added to claim 1).   Although the examiner asked for the amendment, Amgen argued on appeal that it wasn’t necessary to overcome the prior art because a separate amendment was sufficient. However, Amgen failed to provide an alternative purpose behind why the examiner asked for such an amendment. The Federal Circuit faulted that failure of alibi.

[I]f Amgen is correct … the Examiner proposed the Examiner’s Amendment for no purpose at all. Such a reading of the prosecution history is at best unpersuasive.

Slip Op. I’ll note here that the Federal Circuit appears shocked that an examiner would propose an amendment for no legitimate purpose.  As a counterpoint, consider that a large portion of examiner rejections are later withdrawn by the examiner upon further consideration.

In any event, a narrowing amendment made fur the purposes of securing the patent triggers a presumption of prosecution history estoppel that would prohibit a claim of equivalents to re-claim the narrowed subject matter.  One way to get around this presumption is to show that the narrowing amendment is tangential to the equivalent in question.  Here, however, the starch used by Piramal is the same starch disclosed in the prior art overcome by the amendment.  As the Federal Circuit has previously written: “an amendment made to avoid prior art that contains the equivalent in question is not tangential.” Intervet Inc. v. Merial Ltd., 617 F.3d 1282 (Fed. Cir. 2010).

Thus, the non-infringement holding of Piramal stands.

39 thoughts on “Opening a Closed Markush Group

  1. 8

    Can someone tell me why the article used the words “primary” patent on the compound and a “secondary” patent on the formulation? Do these adjectives have any meaning?

  2. 7

    “Slip Op. I’ll note here that the Federal Circuit appears shocked that an examiner would propose an amendment for no legitimate purpose. As a counterpoint, consider that a large portion of examiner rejections are later withdrawn by the examiner upon further consideration.”

    I would like to know what this “large portion” is. Because I withdraw a literally tiny amount of rejections. I’m not even sure I withdrew any rejections in all of last year.

      1. 7.1.1

        JESUS CHRIST!

        Also as a side note I would like to thank MM kindly for his hard work. Through my exploitations of him and his fellow workers I’m up literally like 8k in 2 days. A humble thank you to MM. Truly MM, your hard work and my capital fit together truly well.

      1. 7.2.1

        It might make a significant difference if one is talking about the % of examiner withdrawn examiner 101 rejections without claim amendments versus the % of examiner withdrawn 102 or 103 rejections without claim amendments? [Responsive to applicant arguments or declarations.]

        1. 7.2.1.1

          Certainly Paul, one would expect some very different rates along the various sections of law.

    1. 6.1

      Hilarious article — Thanks!

      … and as I have previously noted, there is more than one “philosophy” that attacks the notion of strong patent protection:

      The Left (by this I mean those hostile to the notion of personal property), and also:
      The Right (by this I do not ((necessarily)) mean the ‘political Right; but instead mean the Corporatacracy of established entities who do not like the notion of others’ personal property that threatens their established position)

      1. 6.1.1

        So true. When it’s their IP – it’s spun gold. A dead company (preferably put out of business by massive infringement) Troll! You’re a troll or worse. And BTW, this NPE label – doesn’t it really boil down to some second class citizen – you don’t get the same rights as a purported practicing owner? Thereby rewarding the big guys for driving the small entity innovator from the marketplace. Thought you might get a kick out of that article.

    2. 6.2

      Is it April Fools Day already? That reads like a parody of the patent huffers and toxic glibertarian gasbags who post here.

      1. 6.2.1

        It’s actually a parody of the Left and their lack of touch with reality.

        Said lack of touch is probably why you feel the way that you do.

      1. 6.4.1

        ?

        His point WAS irony and THAT was what the link provided.

        You really have a bad case of confirmation bias blindness…

  3. 5

    “Amneal’s binder includes HPMC as recited in the claims, but also includes polyethylene glycol (PEG). The major question in the case here is whether Amgen’s Markush Group is “open” or “closed.” In other words, does the presence of the additional compound in the binder (PEG) avoid infringement.”

    Actually, Amneal’s use of Opadry as a binder was not dependent on claim construction of the binder Markush group (“The court’s remaining findings regarding Amneal and the other defendants do not depend on its constructions of the Markush groups and are separately addressed below.”). Since PEG was not alleged to be a binder, the limitation was literally met if Opdary contains 1-5% HPMC (“By containing Opadry, Amneal’s formulation necessarily contains HPMC. HPMC is a binder listed in the binder Markush group of claim 1, so, provided that Amneal’s formulation contains from about 1% to about 5% HPMC, irrespective of whether PEG is present, the formulation literally meets the binder limitation of claim 1.”).

    The major question in Amneal’s case was actually whether the use of an unlisted disintegrant, in addition to crospovidone, which is listed in the disintegrant Markush group, literally read on the “consisting of” disintegrant limitation.

  4. 4

    As a counterpoint, consider that a large portion of examiner rejections are later withdrawn by the examiner upon further consideration.

    Hahaha, clearly written by someone who hasn’t done enough prosecuting. Why have prosecution history estoppel at all then? The only time there would be estoppel would be in the rare malpractice situation where someone admits that a rejection was proper, and even then you wouldn’t need estoppel, you’d have admission.

    Regardless, you’re misapplying who generates estoppel. Estoppel (as the name suggests) is generated by the *applicant*, not the office. It isn’t that the Examiner asked for a narrowing limitation, it is that the Examiner proposed a narrowing limitation *that was acquiesced to by the applicant*. The question isn’t whether the office makes mistakes, the only question is whether the office, in making the demand, intended to limit; and further whether the applicant agreed to the intention.

    This is done for the benefit of all applicants. If you take away the ability of the office to get the limitations it wants by bargaining (which is what Dennis implicitly suggests by rendering all non-litigated demanded limitations null and void) then it forces the office to get what it wants by demanding. And lest you think the office cannot demand, keep in mind that the office is the first mover of a rejection, the first level of review of the rejection, the second level of review of the rejection, and that the third level of review doesn’t really know what it’s doing anyway, and that the timer is always ticking on the applicant.

    tl;dr – If your claim says fruit and I want it to say oranges, either I can ask or bargain with you to make it oranges, which is quick and painless; or you can tell me I can’t make suggestions, and I can fling rejections at you until you figure out yourself, with no help from me, that it needs to be oranges. Either way, if I care about it enough your claim will not issue until 1) it says oranges, 2) you convince me I don’t need to care that much.

    1. 4.2

      Also RG, note that there are not that many examiners sua sponte suggesting specific claim amendments. When they do they may well be doing the applicants a favor by hinting that they could allow the application if so amended. In some cases by distinguishing prior art or in some cases by pointing out what they think may be patentable subject matter in the specification that is not being claimed. It may lead to a useful interview negotiation. It should be more encouraged. But I’m told there are too many applicants these days, including attorneys for foreign applicants, who won’t even take a hint and persist with ridiculously broad originally filed claims that cost their clients high avoidable RCE fees.

      1. 4.2.2

        Also RG, note that there are not that many examiners sua sponte suggesting specific claim amendments.

        Junior examiners have to clear a potentially allowable scope with a primary, so a natural catch-22 is generated: It has to be narrow enough to get the primary to sign off but broad enough for the applicant to take. Threading that needle requires uncompensated work – if you research it enough that your primary might say yes and one of the two balks, you just did a lot of work for nothing, and every counter proposal requires time to explain things to people. So it doesn’t make a lot of sense for juniors to take the initiative even if they see something which might be promising.

        Primaries have fewer procedural hurdles, but because they have less time per case they’re the most likely to just conveyor-line the application and let it develop to the point where they don’t need to spend time going back and forth on multiple issues. Primaries are also less likely to truly believe one particular point of novelty actually exists.

    2. 4.3

      Each of Random, Malcolm and Paul seem to be misreading what Prof. Crouch actually stated (emphasis added):

      As a counterpoint, consider that a large portion of examiner rejections are later withdrawn by the examiner upon further consideration.

      To this statement — about rejections — it appears first Random jumps to:

      Hahaha, clearly written by someone who hasn’t done enough prosecuting…

      Regardless, you’re misapplying who generates estoppel…

      (While also getting the bit about malpractice seriously wrong)

      I do not think anyone was suggesting anything like taking away any power of dialogue and suggested amendments.

      1. 4.3.1

        Random jumps to:

        I’m not jumping to anything, the full paragraph is I’ll note here that the Federal Circuit appears shocked that an examiner would propose an amendment for no legitimate purpose. As a counterpoint, consider that a large portion of examiner rejections are later withdrawn by the examiner upon further consideration.

        Those two sentences are logically irrelevant to each other in terms of estoppel law. When the Examiner proposes something, the Examiner *believes at the time they are making it* that the amendment is necessary, which is a legitimate purpose. The fact that they are later withdrawn does not render the initial suggestion an illegitimate purpose and it *is not a counterpoint* because estoppel depends upon the applicant’s response to the examiner’s motivation, NOT whether the Examiner is ultimately correct. Whether the Examiner is correct or not has zero bearing on the estoppel issue. The only question is whether the Examiner offered a deal that the applicant accepted. Just to be clear, an Examiner could cite something that, by law, isn’t prior art and then say “I’ll withdraw the prior art rejection if you do X” and if the applicant does X, X creates estoppel, because applicant “traded” the performance of X for resolving the dispute over the rejection, even if he think that he could have ultimately won the dispute without the trade anyway.

        I then “jumped to” applying Dennis’ sentences as he actually suggests they be applied – That the underlying correctness of the rejection is relevant. In this case, there is rarely such thing as estoppel. There is only true necessity (an act that overcomes a proper rejection), which exists regardless of estoppel, and non-estoppel (any act in response to an improper rejection). The only time estoppel would exist under Dennis’ sentences’ formulation of the law is in the extremely rare instance where 1) the rejection was incorrect but 2) the applicant states on the record they believe the rejection to be correct and 3) the applicant complies with the term. Because (2)-(3) is essentially malpractice (it surrenders the applicant’s rights due to counsel’s mistake of law, I suppose your implication that this is not per se malpractice is correct), it would never exist.

        I then “jump to” the natural result that flows from such an outcome – since all examiner-suggested amendments would be considered non-estoppel and therefore not true limitations because equivalents applies, the end result would be that there would not be examiner suggested amendments (since legally making or not making the deal generates the same effect). The natural result of cutting out examiner suggested amendments is pretty universally bad for the applicant.

        I do not think anyone was suggesting anything like taking away any power of dialogue and suggested amendments.

        No, he suggested that because Examiners withdraw rejections (something that unquestionably occurs) that an argument that *this* rejection is wrong and therefore estoppel should not apply *here* has merit. But every examiner amendment is subject to the same estoppel nullification attack, so the only way to make an Examiner-suggested amendment “stick” is to demonstrate on the record why it is necessary. But demonstrating on the record why an amendment is necessary is just normal rejection. The whole reason I bother to bargain with you is to avoid having to take the time to write the rejection and argue why I’m correct. If you force the rejection/remarks response writing anyway, now the bargaining becomes superfluous, I’m not going to reject something and then bargain against myself, I’m just going to reject and let you decide what you want to put forward and perform an analysis on that.

        1. 4.3.1.1

          Just because you don’t understand the point does not make it logically irrelevant.

          How the F are you an Examiner?

  5. 3

    Yet, how can a closed Markush (utilizing “consisting” of instead of “comprising”) be anything but . . . closed?

    Does this decision mean that — what — one needs to add a word like, say, “exclusively” to the limitation (e.g. ” . . . selected from the group consisting exclusively of . . . “) in order to make it absolutely clear that “consisting of” really and truly means one (or in rare cases some other explicit number of) from the group listed in the limitation?

    Does this decision mean that “consisting of” does — or in some cases may — actually mean “comprising?”

    Something doesn’t seem right here.

    1. 3.1

      No Markush group is being broken. The claim does not require that the only binder in the accused product to be one of those listed in that Markush group, rather it requires that at least one of the binders in the product must be one from that group in the proportion indicated.

      1. 3.1.2

        No Markush group is being broken. The claim does not require that the only binder in the accused product to be one of those listed in that Markush group, rather it requires that at least one of the binders in the product must be one from that group in the proportion indicated.

        Clearly correct. But on the facts (and this may just be me not knowing the subject matter) isn’t the infringing compound, in actuality, a blended single binder?

        In other words, if a claim “comprises” a basket that “consists of” an apple, a system that has an apple in a basket and a banana is infringing when the banana is carried outside of the basket and not infringing when the banana is placed inside the basket, correct? And isn’t the compound at issue a banana-in-basket situation?

        1. 3.1.2.1

          Adding a banana [or anything else extra] to a claimed basket [not limited to bananas] does not cause or avoid infringement, and note 2 below – this is not just a claim to a binder.

          1. 3.1.2.1.1

            Adding a banana [or anything else extra] to a claimed basket [not limited to bananas] does not cause or avoid infringement, and note 2 below – this is not just a claim to a binder.

            That’s untrue, because the claim requires a basket that *consists* of an apple, so adding a banana to the basket means you no longer have a system that meets the basket element.

            e.g. Claim 1 is a system comprising: a banana, and a basket consisting of an apple. I carry around a basket containing an apple and a banana. I do not infringe because while I hold the first element, I do not hold the second element of the system – a basket that consists of an apple.

            But if I take the banana out of the basket, I now infringe, as I hold both the first and second elements.

            If I keep the banana out of the basket and now add an orange to the basket, I again do not infringe, because while the claim comprises, and therefore allows for additional things, the basket consists, and therefore does not allow for the orange. I have not met the second element of a basket consisting of an apple.

            If I later pick up a second basket, and am now carrying 1) a banana, 2) a basket with an apple and an orange and 3) a basket with an apple, I again infringe the claim. Because (2) is an allowable additional element under the comprising language, and (1) and (3) are the first and second elements of the claim, respectively.

            What I view the dispute in this case is that PEG is the “banana” of the claim. If you view PEG and HPMC as part of the same thing – a blended PEG/HPMC binder (i.e. the banana is in the basket) – then there is no infringement because the close markush group element is not met. On the other hand, if you view the banana as being outside of the first basket or in a second basket (there is a PEG binder and a HPMC binder) then there is infringement because the closed markush of HPMC binder is met, and the comprising term allows for a second binder. But what I am asking is, as a factual matter, don’t the PEG and the HPMC interact such that you have to consider them being “in the same basket”? What you have is the function of binding that is performed by the combination of PEG and HPMC?

    2. 3.2

      The Markush group is only “closed” in the sense that at least one of the listed components must be present. It does not mean that another component outside the group must not be present. Otherwise, the “comprising” transition word would be turned into “consisting of”.

  6. 2

    In claim 1, after the “comprising,” there is a non-Markush limitation and two other Markush group limitations besides the one discussed.

  7. 1

    I’ll note here that the Federal Circuit appears shocked that an examiner would propose an amendment for no legitimate purpose. As a counterpoint, consider that a large portion of examiner rejections are later withdrawn by the examiner upon further consideration.

    If the presumption of validity is to be relied upon, then it’s contradictory to also rely upon the assumption that examiners sometimes do things for no legitimate purpose.

    1. 1.1

      Can’t quite get to your location regarding a granted property right ‘stick in the bundle’ and a prosecution step that may occur pre-grant.

      The particular stick in the bundle is the transformation of that personal inchoate right into being a full legal right. It’s more than — and not strictly dependent on — any one prosecution action.

Comments are closed.