Functional Claim “Raises the Bar for Enablement”

by Dennis Crouch

Amgen v. Sanofi and Regeneron (Fed. Cir. 2021)

The decision here provides another anti-functional-limitation decision — this time rendering Amgen’s monoclonal antibody claims invalid.  Although enablement decisions traditionally differed greatly between biotech and computer innovations, the decision here relies heavily on the Federal Circuit’s recent software decision in McRO, Inc. v. Bandai Namco Games Am. Inc., 959 F.3d 1091 (Fed. Cir. 2020).

The first jury sided with Amgen, the patentee, finding that Regeneron had failed to prove that the asserted patents lacked enablement or written description.  On appeal, the Federal Circuit vacated the resulting final judgment and ordered a new trial based upon errors in evidentiary rulings and jury instructions.  On remand, the second jury again sided with Amgen.  The district court though favored the defendant and awarded Judgment as a Matter of Law (JMOL) for lack of enablement. On appeal, the Federal Circuit has affirmed.

[Image of second verdict above]

The patents at issue claim monoclonal antibodies used to help reduce LDL-C levels. US8829165 and US8859741.  Amgen’s product is evolocumab (Repatha) and Regeneron’s product is alirocumab (Praluent).  Both drugs have been shown to be effective and each sell for around $14,000 per year.  In the lawsuit, Regeneron admitted to infringement, but challenged the patent’s enablement — arguing that the claims extend well beyond what was taught in the specification.

Anyone reading the claims will quickly notice that the claims are claimed functionally — the claims are basically directed to all monoclonal antibodies that bind to a particular portion of the protein PCSK9 and consequently block PCSK9 from binding with the LDL-C.

An isolated monoclonal antibody [that] binds to [one of the residues of PCSK9] and … blocks binding of PCSK9 to LDRL

‘165 patent.   Sanofi argued to the jury that the claims encompass millions of potential antibody candidates and that the specifications lack sufficient guidance to allow a person of skill in the art to practice the full scope of the claims without undue experimentation in the form of substantial trial and error. Amgen provided contrary testimony: “The jury heard evidence from which it could conclude that the number of distinct antibodies within the claims was around 400.”  After hearing both sides, the jury found the claims enabled. But, Judge Andrews rejected the jury verdict and sided with the patentee — holding that no reasonable jury could have found otherwise.

Enablement: The patent act requires that the patent specification “enable any person skilled in the [relevant] art . . . to make and use” the claimed invention.  35 U.S.C. 112(b).  Like obviousness, enablement is a question of law based upon a collection of factual findings.  The Federal Circuit generally considers the Wands factors in enablement cases — particularly when asking whether the gaps in disclosure require undue experimentation.  The Wands factors:

  1. Quantity of experimentation necessary: When more experimentation is needed, it is more likely undue experimentation;
  2. Amount of direction or guidance presented in the specification: When less guidance is provided, it is more likely that practicing the invention requires undue experimentation
  3. Presence or absence of working examples: Their absence suggests undue experimentation;
  4. Nature of the invention: This one is basically meaningless, but it generally is seen in comparison to the level of prior art;
  5. State of the prior art: If the invention is a major step beyond the prior art, then it more likely requires undue experimentation;
  6. Relative skill of those in the art: If PHOSITA is relatively low skill then it is more likely that undue experimentation is needed to practice the invention;
  7. Predictability or unpredictability of the art: Unpredictable arts are more likely to require undue experimentation in order to practice the invention;
  8. Breadth of the claims: Broad claims are more likely to require undue experimentation to practice the full scope of the claims.

In re Wands, 858 F.2d 731 (Fed. Cir. 1988).

The functional claim language and resulting potential for so many compounds fitting within the claim scope focus attention on the final factor — claim breadth. Although Amgen appears to have provided guidance and framework for creating and testing the efficacy of each of its claimed compounds, the problem is that there are many thousands of potential compounds — only some of which will actually work effectively as claimed.

While functional claim limitations are not necessarily precluded in claims that meet the enablement requirement, such limitations pose high hurdles in fulfilling the enablement requirement for claims with broad functional language. . . .

[T]he scope of the claims is broad. While in and of itself this does not close the analysis, the district court properly considered that these claims were indisputably broad. . . . If the genus is analogized to a plot of land, the disclosed species and guidance “only abide in a corner of the genus.” AbbVie (Fed. Cir. 2014). Further, the use of broad functional claim limitations raises the bar for enablement, a bar that the district court found
was not met.

Slip Op.

Although the court identifies the limitations in the claim as “functional” rather than structural. These particular functional limitations are important because they direct researchers in how to go about identifying compounds covered by the claims.  In other words, its not surprising that researchers would say that their invention is compounds that bind to PCSK9 — because that is exactly how they have been screening for the compound.  Amgen suggested a randomization-and-screening roadmap as a pathway for figuring out which antibodies would bind with PCSK9.  On appeal though the Federal Circuit held that that the the amount of experimentation was too much.

The functional limitations here are broad, the disclosed examples and guidance are narrow, and no reasonable jury could conclude under these facts that anything but “substantial time and effort” would be required to reach the full scope of claimed embodiments.

We therefore conclude that, after weighing the Wands factors, the court did not err in concluding that undue experimentation would be required to practice the full scope of these claims.

Slip Op.   Note, the court took care to suggest that it may be possible to enable an entire genus, but only perhaps. “We do not hold that the effort required to exhaust a genus is dispositive. It is appropriate, however, to look at the amount of effort needed to obtain embodiments outside the scope of the disclosed examples and guidance.”

In this case the burdens are important.  The jury found that the defendants had not met their burden of clear-and-convincing evidence of invalidity.  The district and appellate courts considered the same evidence and flipped — holding that the only reasonable outcome that a jury could reach is that the claims were proven invalid with clear and convincing evidence.

Amgen argued that that the potential breadth of the claims was “artificially inflated” by Sanofi; that the “millions” number was contested on the facts; and “A reasonable juror could easily have rejected Sanofi-Regeneron’s argument [as not] clear and convincing.”   On appeal, the Federal Circuit notes that they are “not concerned simply with the number of embodiments but also with their functional breadth,” although the court later appears to side with the defendant in concluding that “the evidence showed that the scope of the claims encompasses millions of candidates.”

This is a big case for enablement and I expect to see a petition for rehearing — likely with substantial focus on the standards for reviewing and rejecting a jury’s factual findings.

= = =

Consider – Greg R. Vetter, Patent Law’s Unpredictability Doctrine and the Software Arts, 76 Mo. L. Rev. 763 (2011) (arguing against the categorical lines on enablement).

82 thoughts on “Functional Claim “Raises the Bar for Enablement”

  1. 8

    A passing thought on an admittedly aging thread (and might need to be raised again next time the issue arises), is that ANY view of “raising” or “lowering” the bar for enablement should not be passed off and blamed on patent applicants.

    What we see with the entire trend towards the “boogeyman” of “functional claiming” is merely a reflection of the predominant form of innovation (Kondratiev Fifth Wave) coupled with the patent profanity induced by the Courts themselves in their effort to increase the power of PHOSITA for the side of the coin of obviousness. Every time you increase that power, you necessarily lower the bar (the opposite of the desired heightening) for enablement.

    This may well be an unintended consequence, but it also is unavoidable, given that what is required for enablement is but the other side of the coin.

  2. 7

    Here is another, prior, Fed. Cir. decision on 112 non-description for a broad pharmaceutical patent claim: University of Rochester versus G.D. Searle & Co., Case No. 03-1304: link to;

  3. 6

    The full scope argument is usually only used for a reverse doctrine of equivalence.

    I haven’t read the opinion yet. Is there a link? This has the foul stench of Taranto to it.

  4. 5

    Much of the argumentation here boils down to: “that claim is really broad, so it would require too much experimentation to practice it across its full scope.”
    Which is silly from a real-world perspective. Why would anyone ever want or need to practice “the full scope” of such a claim? As a competitor I wouldn’t have to make millions of such antibodies – a few hundred would be realistic, and just one would be enough.
    We shouldn’t be musing whether it would take undue effort to make every embodiment of such a claim – of course it would. We should be asking whether the patent enables the reasonable, real-world practice of the invention, by ordinary, skilled people, under conditions that are typical for that field.

      1. 5.1.1

        Absolutely – as has been noted by several (including yours truly).

        Worth also noting — the vast majority of patent claims fall into the ‘category’ of 101’s ‘improvement thereof.’

        Does any (single) instance of an improvement patent NEGATE the patent landscape upon which the improvements exist?

        To borrow from Random’s example, the claim, “I claim a bow,” would be proper to him, and yet — at the same time — would be IMPROPER once someone comes along and claims an improvement (such as I claim a bow with a compound mechanism at the bow string to bow body interface). Clearly, the original inventor of “the bow” did not enable across the full scope of ALL bows.

        And down the rabbit hole goes Alice.


          “Worth also noting — the vast majority of patent claims fall into the ‘category’ of 101’s ‘improvement thereof.’”

          Then why arent the claims drafted as improvement patents, as CFR requires (or suggests, depending on your reading of the word ‘should.’)



            No wait,

            You mean, you want people to actively choose the Jepson format?

            How long have you been an examiner?


              Not an examiner, and have never been one.

              But if you say “My invention is an improvement,” and your claims dont particularly and distinctly point out what the “improvement” is (vs. what is old), then there is a colorable argument that you have not “particularly” and “distinctly” pointed out what you consider your “invention” (ie., you have not particularly pointed out what the improvement is) as required by 112.

              which I assume is the statutory basis for the CFR. Since you like statutory interpretation, what do you take the words “particularly pointing out” and “distinctly” to mean in 112(b)?


                Whether one “says” my invention is an improvement and whether it IS an improvement are two very different things (plus, Jepson is an option, and not required as you appeared to intimate).


                The invention — as a whole — is described by the claim.

                There is zero actual legal requirement to use Jepson format in the US Sovereign.


          (But yes, you are right there is a problem with existence of ‘improvement claims’ and the idea of enabling the full scope of the invention. This is a conversation that I think was had on week 5 of my first ever patent class in law school. So deep stuff here, anon )


          To borrow from Random’s example, the claim, “I claim a bow,” would be proper to him, and yet — at the same time — would be IMPROPER once someone comes along and claims an improvement (such as I claim a bow with a compound mechanism at the bow string to bow body interface). Clearly, the original inventor of “the bow” did not enable across the full scope of ALL bows.

          I certainly agree that if the claim was “I claim a bow” that claim is invalid. That’s literally what happened in the Morse case, where Morse attempted to claim all remote printing because he had the telegraph. Obviously he didn’t invent the fax machine (the compound bow of this analogy), and the court didn’t call the claim valid or even wait around until the fax machine was invented to prove the invalidity of the claim – they simply reasoned that the failure to foreclose new inventions (such as the fax machine) was sufficient to render the claim invalid even before they came about. So when you say “‘I claim a bow’ would have been proper, but an improvement would show it to be improper” that’s because you lack understanding of precedent – the claim to “a bow” is invalid in the first instance. Though I don’t hold it against you, as often the federal circuit needs a non-posited embodiment sitting right in front of them (Lizardtech for example) to make the mental leap the supreme court makes unaided.

          But the person who invents the bow doesn’t claim a bow. The bow is the terms that arrives after the invention to describe mentally grouped structures. We call a compound bow and a bow “bows” but we don’t call a catapault a “bow” even though it also uses string tension to propel projectiles. The initial claim for a bow wouldn’t be the words “I claim a bow” it would be a structure that claims in detail the (1) curved wood, (2) tensed string and (3) how the tensed string interacts with the curved wood (and probably 4, a projectile). You obviously can’t use a bow without the string tension (as that propels the projectile), nor the wood (as that supports the tension), nor the interaction (as simply putting a string and a piece of wood next to each other does nothing). Consequently you need 1, 2 and 3 at a minimum for enablement. But because a compound bow has an entirely different (3) (and marginally different (2)), it simply doesn’t read on a proper bow claim.

          This is a silly argument you’re putting forth – are you suggesting that the structural difference between a normal bow and a compound bow is inarticulable? Or is your argument that your instinct to abstract is so strong that you can’t fathom that the base bow inventor would include limitations that would ultimately articulate that difference? That’s kind of the point of dependent claims – you may have an independent claim that proves to have been overbroad, but dependent claims provide the proper scope. Or conversely, you could simply avail yourself of 112f and not have to worry about this kind of validity and instead argue over whether the compound bow is an equivalent infringement.

          Either way it’s pretty clear that the inventor of the base bow isn’t the inventor of the compound bow, so I don’t know under what legal theory you think he should get a valid claim that covers it? If the base bow is so unoptimized that it can be substantially improved upon during the patent term, the system is supposed to provide incentive to the compound bow inventor to come forth. But if the base bow claim validly covers the compound bow, the system provides no incentive to the compound bow inventor to disclose. There is no mechanism where the person who invents the base bow gains patent rights over the compound bow when the art would recognize a nonequivalency between the two, that is the system functioning as intended.

          if we consider the open-ended claim transition “comprising.”

          Comprising as a transition isn’t a problem, but using comprising in the definition of elements or in “wherein” clauses begs for a huge unforseen problem to crop up. The comprising limitation must be (1) necessary and (2) nearly infallible if you use comprising in that situation, because if someone challenging can articulate a whole class where the limitation fails, the claim is unenabled.


            You created a strawman, then tried to say that my argument is your created Strawman.

            My argument is not your created strawman.

            You have not addressed the point that I have presented and the fact that improvement patents — under your ‘theory’ — necessarily show that the prior patent was “improper” for not covering the full scope.

            An important aspect here (and one you may or may not grasp) is that there is a reason why patents are Negative rights and that they are not rights to affirmatively make something. This IS because an improvement patent may well ‘infringe’ on a preceding patent if one were to ‘make’ the object of the improvement patent. The actuality of scope overlap implicate your ‘theory’ on FULL scope coverage.

            While you like to use your example of ‘bow’ and ‘gun,’ the real world does not have rungs on the Ladder of Abstraction as far apart as you try to use with your example. When you try to apply your example and your ‘logic’ in the real world, in which the Rungs are closer together, your logic falls apart and necessarily defeats the very reason for having a patent system in the first place.

            This happens to coincide with past criticisms of mine of your views that lead to only two type of innovation: the Flash of Genius and the “Oopsi” types.


              You have not addressed the point that I have presented and the fact that improvement patents — under your ‘theory’ — necessarily show that the prior patent was “improper” for not covering the full scope.

              Again, the validity, unsurprisingly, depends on what the original claim said. The original claim obviously didn’t say “I claim a bow.” The original claim, in order to be enabled, would have to describe the structure of the bow which would include the detail of how the bow and the string interact. Thus the scope of the initial claim wouldn’t cover the compound bow, so the later discovery of the compound bow would not be evidence that the initial claim is overbroad.

              If, theoretically, one wrote the claim for the initial bow to not include this feature and one nonetheless considered it enabled at the time (I don’t know how you would consider it enabled when it does not describe the operative feature that causes the bow to fire, but whatever), then the claim still does not *become* unenabled when the compound bow is presented. The invention of the compound bow does not change the scope of the base bow claim, nor does it change the scope of the base bow disclosure. The claim was always too broad. Ignoring features of your invention always poses some level of threat of rendering the claim invalid (either from 112 or 103). That’s why dependent claims exist.

              This IS because an improvement patent may well ‘infringe’ on a preceding patent if one were to ‘make’ the object of the improvement patent.

              Well initially let me say that you just don’t understand what improvements are. A compound bow is not an improvement on a bow. A compound bow is a different invention. So is a gun. They are non-equivalent in terms of structure, use and result. An improvement patent would be a patent on a rubber grip for a bow.

              Beyond that, by definition your improvement must be novel and nonobvious over the parent invention (or you wouldn’t get a patent) so there has to be language to put in the claim, and there is nothing gained by including the prior art in your claim.

              I mean your logic is pretty circular – you’re trying to prove that functional claims are valid by stating that improvement claims must necesarily infringe on their base, but neither of those are true. I suppose if you articulated your rubber grip as a system comprising a bow and the structure of the rubber grip, then you are correct that your invention inherently would have to infringe the bow patent. But the bow adds nothing to the patentability of that claim – you could simply articulate the structure of the rubber grip in the claim, and state in the specification that the utility is an improvement to comfort in use with a bow. Conversely, you could write a jepson claim that you dislike so much and place the bow in the preamble. Either way, there’s never any requirement that your claim include prior art contextual limitations – your claim doesn’t have to demonstrate utility, it just has to have utility.

              Similarly, you wouldn’t claim a compound bow (as you didn’t invent the bow) you would claim a cam system interacting with string affixed to endpoints and state in the specification that one utility of this would be to increase the tension applied in a bow context. You can sell a base bow without a cam system, and you can sell a cam system without a base bow, so there’s no conflict. You only generate a conflict when someone tries to claim things they didn’t invent – either the first inventor claiming a function or the second inventor needlessly claiming the other features of a bow they knew were in the prior art.

              This is not only ideal, but in the past was REQUIRED – it was an 112b aggregation rejection to include the prior art bow features in the compounding feature claim because the inclusion of the prior art features failed to point out and particularly claim what *this* invention was. That requirement, in a post-52 world, was relaxed on the CAFC level (though one questions if the supremes will ever revive it).

              When you try to apply your example and your ‘logic’ in the real world, in which the Rungs are closer together, your logic falls apart and necessarily defeats the very reason for having a patent system in the first place.

              You act as if the applicant has only a single claim to work with. But the applicant has infinite claims of infinite depths with which to work. The correct way to do something is to take the picture and then explain why each limitation that is being stripped from the picture is supported by the knowledge in the field or the specification. In other words, rather than immediately jumping to the most abstract thing and working top-down, you work bottom-up and each inductive leap (what you call an abstraction) taken from the picture is explained and supported, and multiple levels are claimed so that if an inductive leap turns out to be wrong, you still have a dependent claim that is valid on the level below it.

              As an example, if the bow you made was made out of yew wood, the attorney should have a claim that your invention is made out of yew. That may very well be a nested dependent claim. If your independent claim simply says “wood” there should be an explanation in the specification as to why any other wood could be substituted for the yew. Unlike most specs I see today, which start with the word “wood” and include a single sentence which says “in one embodiment the wood is yew” it should be written that in one embodiment the wood is yew, but any wood which can be bent without disforming or cracking when X amount of force applied to it can be used, as that is the operative feature of the wood in this device. Example woods that do so are types A, B, C and D, but quite possibly any type of wood could be used. The inventor posits using any type of wood instead of the yew.” So when it turns out that cedar cracks and is unenabled you don’t tank the entire claim set – you may be enabled at claim 3, which limits the wood to non-cedar types A B C or D, or claim 5, which makes the wood yew. The “problem” you have simply doesn’t exist, it only appears to exist because prosecutors are more interested in flexing their abstraction muscle than actually imparting supporting knowledge to artisans.

              I don’t know why you’re having such trouble here. You admit (even if you deride it) that picture claims could be made. You know abstractions can be made. You certainly should be aware that dependent claims exist. And you’re commenting on a post where a claim was invalidated due to a scope of enablement failure. The solution seems self-evident: You need to support every abstraction that is made from a picture claim or you run the risk that you’ll hit a scope of enablement problem. The fact that you don’t want to doesn’t mean it can’t be done.


                I am not the one having trouble understanding the problem with your attempt to misunderstand what improvement claims are and how they wreck your desired “logic” about “full scope.”

                You still have not addressed the point that I presented and you still attempt to re-characterize (and mis-characterize) what I stated.

                I used your “I claim a bow” – -> “I claim a gun” mantra, and now you want to “revise” the bow claim in order to avoid the outcome of your own logic.

                But you are applying your own changes only in one direction – and that be the wrong direction.

      2. 5.1.2

        Dennis, you can get to the same issue you have even with “consisting of”
        as a transition.

        1. A car consisting of a body and four wheels.

        2. The car of claim 1, wherein the body is constructed of unobtanium.

        Claim 2 is non-enabled (assuming unobtanium cant be made). Since it is narrower than claim 1, claim 1 is also non-enabled.

        No “comprising” necessary. There are obviously limits on meaning of “full scope.” Just very hard to elucidate what they are.

      1. 5.2.1

        A bette question**: would we want that?

        ** along the lines that your question reflects a certain type of Patent Profanity that once imbibed, tends to multiply…

        …like rabbits.

      2. 5.2.2

        They could have just argued for 112(f) construction, in which case they would protect the disclosed antibodies and equivalents. If antibodies are all equivalent, there isnt a problem here.

        Why reinvent the functional claiming problem? We already have an answer.


          You do of course realize that 35 USC 112(f) is for PURELY functional claiming (as opposed to the Prof. Crouch coined term of Vast Middle Ground), eh?

          That “answer” you seem to seek – much like the ‘answer’ of Jepson format – is NOT the answer that you think it to be (leastwise from an actual practitioner’s perspective).


          They could have just argued for 112(f) construction, in which case they would protect the disclosed antibodies and equivalents. If antibodies are all equivalent, there isnt a problem here.

          The fact that every claimset which accurately discloses at least one embodiment (and unfortunately, that number is not 100%) doesn’t include at least one independent formatted in means-plus just blows my mind. Talk about bad prosecuting.

      3. 5.2.3

        “Could the patentee have included language within the claim to limit the scope to something reasonable, as you suggest?”
        I don’t think they should have to – doing so may make the claim narrower but not necessarily more enabled. Though I share the discomfort with very broad claim scope. Maybe to control claim breadth, it would be better to rely on written description and prior art. But for enablement my proposition is, to paraphrase: “the question is whether a skilled person could practice any one embodiment within the claimed genus without undue experimentation, not whether it would be too much work to practice ALL embodiments.”


          the question is whether a skilled person could practice any one embodiment within the claimed genus without undue experimentation

          So by your view, prior to written description being started by the federal circuit in like 1990, we had almost 200 years where someone could have invented one new invention and claimed “1. All novel, non-obvious subject matter that came into existence after the filing date” and that scope would have been enabled because their one disclosed embodiment was enabled?

    1. 5.3

      A method for treating Alzheimer’s comprising administering an effective amount of a pharmaceutically acceptable ingredient that reverses the effects of plaques which comprise of abnormal clusters of protein fragments built up between nerve cells, and tangles which comprise dead and dying nerve cells comprising twisted strands of proteins.

      My only example in the specification is a small molecule compound X.

      Can I use this claim to prevent a company from selling an antibody vaccine that reverses plaques and tangles?

    2. 5.4

      It’s not about practicing the full scope of the claim
      Its about having a monopoly– and thereby preventing others from practicing– the full scope of the claim.

      If you want to get a monopoly over the full scope, you have to enable the full scope.

      If you want a more limited monopoly, then you dont have to enable the full scope. This claim would have been saved by arguing 112(f) should apply, in which case it would be limited to the actual disclosed antibodies *and their equivalents*. If the competitor’s antibodies were simply equivalent to what you disclose, no problem.

      1. 5.4.1

        Thanks ksksksks for addressing the real 112 enablement issue for claims re broadly functional or end-result claims, especially in the “unpredictable arts.”

    3. 5.5

      “Which is silly from a real-world perspective. Why would anyone ever want or need to practice “the full scope” of such a claim?”

      Perhaps you should ask the patentee why they wanted that broad scope.

    4. 5.6

      At present, claim scope is not limited to “reasonable, real-world practice of the invention, by ordinary, skilled people, under conditions that are typical for that field.” It seems inequitable for infringement liability to be broader than the enablement requirement.

      Wouldn’t narrower claims also solve this problem?

    5. 5.7

      All cars prior to my invention only achieve 50 miles per gallon of gas, maximum.

      I make a car that achieves 51 mpg.

      “I claim a car having fuel efficiency exceeding 50 mpg.”

      I have now covered a car having a fuel efficiency of 100 mpg. But I have no clue how to make that. According to you, the claim is enabled.

      1. 5.7.1

        I think that open-ended ranges are a bit of a non-sequitur here.

        Such certainly do as you indicate, but I think that a different point was being attempted to which you are offering the open-range counter example.


          The open-ended range thing is a bit of a conundrum. Its interesting how a range bound on one side raises red flags for practitioners, but an unbounded range doesnt for whatever reason.

          1. A car comprising 4 wheels.

          2. The car of claim 1, wherein the car has a fuel efficiency of at least 51 mpg.

          Its amazing how many practitioners will flag claim 2 as being potentially non-enabled because it has an ‘open-ended range’ (51 to infinity), when in fact claim 1 has an even larger open ended range (zero to infinity). That is: if claim 2 is non-enabled, then claim 1 is too since its broader than claim 2. But for some reason (I know, that resistor case), practitioners flag claim 2 but not claim 1. The *real* rule must not be that open-ended ranges make something non-enabled– because every single claim encompasses an open-ended range– but something about if the cited boundary is the point of novelty or something.


            Do you have a statutory “tie” to this “point of novelty” thingie?

            (the next question might then be to compare that tie to the ‘claim as a whole’ view)


              I was mostly just pointing out that, practically and logically, the rule can’t possibly be that “open ended ranges are non-enabled” (even though so many practitioners think thats the rule). If you look at the cases where its been applied, its generally because the bound end of the range itself was added to overcome novelty/obvious rejections (hence the suggestion that the rule is really some type of point of novelty analysis, though the courts havent set that forth yet).

              If I had to come up with a statuory tie, its that 112 requires teaching how to make and use the “invention.” The “invention” is defined separately from the “claimed invention” in 201, and we have seen repeatedly in 101 analysis that the Courts treat the “invention” as essentially a point of novelty analysis to determine what underlies the “claimed invention.” Thus, where statute uses the term “claimed invention” you look at the claims as a whole. Where statute uses the term “invention” you look to what was actually invented or discovered that underlies the claims (i.e., point of novelty analysis). Again, based on separate definitions in 201. (I know you wont like this, because you think they should always look at claims as a whole, but it is a statutory justification).


                Any Court use of “Point of Novelty” for 35 USC 101 is of course ultra vires.

                Please tell me that you at least recognize that a broken scoreboard won’t be telling you the correct score.


            …and of course, an underlying push (and pushback) is, “just how exacting of a picture claim is necessary to be a proper claim.”

            “Comprising” is just one manner of using the Ladders of Abstraction (which illuminates the fantasy of “Exact Picture Claim” claiming.

            Terms sounding in function (distinct, mind you, from PURE functional claiming), are also tools used in the art of claim drafting that reflect the notion of Ladders of Abstraction.

            A blast to the past: Malcolm Mooney was one such “exacting” picture claim kind of guy, and he would never have an answer when the issue arose as to what was the exact physical structure of a claim with the word ‘comprising.’ Even as he would expound on his feelings, he recognized that his feelings were not (and could not) be the law.

            This ties back to the basic nature of a patent right being a negative right (and why that is so).

  5. 4

    Dumb decision. Every immunologist knows how to make antibodies (inject a mouse/rat/rabbit/goat etc. with a hapten having the structure of interest and an adjuvant to elicit an immune response) and to screen them for function. The more functions a given antibody has to have, the fewer members of the genus.

    There were claims that recited particular amino acid sequences are part of the characterization of the antibodies. Were these not asserted, were they admitted to be not infringed, were they also found to be invalid?

    1. 4.1

      @Atari man – I have noticed the trend is only to allow exemplified antibody claims. Meaning, (like you point out in your second paragraph) those claims to the epitope sequences. I also was confused about Amgen’s other claims that did recite aa sequences. I wonder if Sanofi/regeneron’s products had different sequences.

      1. 4.1.1

        To follow-on my own comment – Is the practical lesson here to file a “genus” functional claim set, then in a divisional or con, file the specific epitope claims so that if the broad functional antibody claims are held invalid, the “specie” claims survive?



            In the pharma space, I would want a claim that is exactly the product. Then I would try to back out form there. If the epitope is 50 aa in length, “a 50 aa sequence having a 90%, 80%, 70% homology to SEQ ID #1” or “a 50 aa sequence comprising a 10 aa contiguous sequence to SEQ ID #1 (which is a specified 10 aa sequnce).” But to back out to a functional claim, i.e, and antibody “y” that binds to an antigen “x” thereby inhibit “Z” is a but of a stretch to me. But I am open to other’s thoughts.


              Would *every* AA sequence having 70% homology work? Would every sequence having the 10-aa contiguous sequence work? Pretty conceivable I could make an AA with that sequence that for whatever reason doesnt bind right (because of folding in other regions, or whatever).

              In which case you are covering under your scope something which doesnt work and is non-enabled because you havent taught how to ‘use it’.

              WHich is to say, ideally, you should still include the function in the claim, along with the structure. And doing so does seem to get ridiculous.

              But if other AA use the same binding sequence, I would say it is equivalent to your disclosed AA. In which case either just using 112(f) functional claiming OR just claiming the specific AA in the claims should cover under DOE. Problem is DOE is underdeveloped.


                Pretty sure the fact that non-working examples may fall into the claim scope do not wreck the claim based on ‘not teaching how to use.’

                Bio/Pharma is not my area, so I do not recall the case citation, but it should be easily provided by those in this area.

                1. The standard is whether the inoperative embodiments can be identified as inoperative without “undue experimentation.”

                  In pharma, identifying whether a given molecule/protein will actually work can be pretty tough and require tons of experimentation, possibly through several clinical trials. Not saying thats ‘undue’ necessarily, but if figuring out how to make antibodies that bind a certain site is ‘undue’, then, well, you can see the problem.

                  Just saying that xtian’s suggestion of structural claiming still runs into “enable the full scope” of the claim problems, it just shifts it from “how to make the full scope without undue experimentation” to “how to use the full scope without undue experimentation” (which in turn requires “how to identify operable from inoperable embodiments without undue experimentation”).

                  The problem is there is no way to actually enable the literal full scope of the claim. Its just courts throw that out there without putting a limiting principle on it.

                2. MPEP 2164.08(b), though unfortunately of course you can always question if MPEP is accurate in its description of the law.

                  But see, for example:
                  29 F.3d 1555 (” As Dr. Goeddel testified, an infinite number of permutations of natural t-PA are covered by these other definitions. Many of these permutations would not be capable of binding to fibrin and would thus be inoperative. There is no basis provided in the specification for determining which of these permutations are operative and which are not. The point is supported by Dr. Larsen’s testimony to the effect that the properties of these permutations were “totally unpredictable.” […] The determination of which permutations are operative would thus require an undue amount of experimentation. Thus, we are unwilling to say that the specification satisfies the enablement requirement of 35 U.S.C. § 112 ¶ 1 (1988) with respect to these broader definitions.)

                  It’s (one reason) why including non-operable embodiments in an application (including the fact that they are non-operable) is advised, and why including a test to separate operable from non-operable species is advised.

                3. Thanks for the leads, but a case in which the presence of inoperable items leading to a failure is not the same as a case in which the presence of inoperable items did NOT lead to a failure.

                  Telling me what I cannot do does not necessarily tell me what I can do.

                  I was looking for the latter.

                4. Atlas Powders 750 F.2d 1569 and citing cases work for you there.

                  “Even if some of the claimed combinations were inoperative, the claims are not necessarily invalid. It is not a function of the claims to specifically exclude . . . possible inoperative substances. . . .” The district court held it would have been impossible for Bluhm to list all operable emulsions and exclude the inoperable ones. Further, it found such list unnecessary, because one skilled in the art would know how to select a salt and fuel and then apply “Bancroft’s Rule” to determine the proper emulsifier.”

  6. 3

    To date, I have published two articles on functional elements. These are shown below. LATE-BREAKING NEWS – – – I will be one of the featured speakers at this year’s AIPLA meeting. The meeting will be a virtual one, and it will be in May 2021.

    Tom Brody (2014) Functional elements in patent claims, as construed by the Board of Patent Appeals And Interferences, 13 J. MARSHALL REV. INTELL. PROP. LAW. 251.

    Tom Brody (2008) Functional elements can ensure allowance of genus claims, 90 J. PAT. & TRADEMARK OFF. SOC’Y. 621.

  7. 2

    Here in Europe, in the “unpredictable arts”, the notion of “plausibility” holds ever-greater sway, when mulling over enablement, sufficiency, novelty and obviousness.

    Example: was it really “plausible” at the filing date, that the scope of the claim in issue is commensurate with the contribution to the art boasted by the Applicant in the application as filed, and relied upon for patentability of that claim.

    Does the notion of plausibility play any role in patent cases in the USA?

    1. 2.1

      Does the notion of plausibility play any role in patent cases in the USA?

      In which section of the statutory law would you think that such would lay?


          Let me rephrase: which ONE section of the statutory law would you think that such would lay?

          Here’s a hint: your answer throwing in both novelty and obviousness are simply not necessary. Or perhaps you can explain why you think that such are necessary…


            For all three sections there is from time to time a need to decide what any given document discloses (to the PHOSITA) and what it enables (for the PHOSITA). The notion of plausibility has become relevant to these deliberations, here in Europe.

            It cannot be, that only Europe has a difficulty, from time to time, deciding what the “teaching” that any given document purports to deliver does in fact deliver.


              You are confusing yourself and conflating separate portions of law.

              Just stop while you are behind.

                1. Can you do one of those broken link thingys that shows your “multiple threads.”

                  [sigh; what a maroon]

                2. Your comment does not seem to be a genuine attempt at conversation, as you use a known fallacy of “broken link” as some type of ‘punch line.’

                  Are you interested in an actual conversation?

                3. Do I need a reason to NOT reply to what I deem a not-actual conversation?

                  You have your ability to choose your responses (well, within limits, although you seem to want to v10late those limits).

                  Why would you seem to want to deny me my ability to choose?

                  And here, my pal Shifty, the cold hard facts of obsess10n bear out MY position.

                  I routinely engage all types of people and speak to and about all types of patent law issues.

                  To date – this year – (other than a mere FOUR posts by “Thomas Pain”), you have a 100% rate to or about me, with a 0% rate of being on topic to patent law issues.

                  and THAT is why YOU are the culpable one – you are the ONLY culpable one.

                4. No. What I asked was Why did you “reply” to what you deem is not an “actual conversation?”

                5. I answered your question with both a question and a discussion.

                  The ball is in your court to choose to reply with something meaningful (or not).

                  It is precisely your propensity to choose the “or not” that I have monitized.

                6. Yes. What I asked was Why did you “reply” to what you deem is not an “actual conversation?”

                7. A “no” and a “yes” followed by the same cut-n-paste statement does not move the ball out of your court for a substantive answer.

                  All it shows is your game-playing.

                8. We disagree with both your premise and your conclusion.

                  We will wait while you wahh wahh wahhh to management.

                9. Thank you for showing the fact that your mere disagreement means nothing in the face of evidence.

                  As to the ‘wah,’ you still are purposefully being obtuse.

                10. So they have some crazy lay convenient definition of “cyberstalking” where you are that you cannot defend under US law.


                11. … aren’t you behind schedule now of dropping in new random obsess10ns to or about me on thread discussions dormant for over a month?

                  What’s the matter? Feeling sleepy?

  8. 1

    Having only read the synopsis here I can already identify a mistake:

    While functional claim limitations are not necessarily precluded in claims that meet the enablement requirement, such limitations pose high hurdles in fulfilling the enablement requirement for claims with broad functional language. . . .

    [T]he scope of the claims is broad. While in and of itself this does not close the analysis, the district court properly considered that these claims were indisputably broad. . . .

    Assuming this is an accurate representation of what was written, the “in and of itself” language is incorrect. A claim to a function not bounded by limitations nominating a particular means inherently violates enablement, written description, indefiniteness, and is abstract. It is the thing you CANNOT DO – You cannot invent one machine that represents one way of achieving a result and claim that you invented achieving the result. One does not invent achieving a result, the result is merely one expression of the use of the structure or process one might possess. One can only invent the means, not the function, result, or utility itself.

    My standard example for this is the person who invents the bow and brings it to his terrible lawyer who (all too commonly these days) tells him that he invented the machine that performs the function of firing a projectile. Even assuming that the bow is a novel expression of that function or utility, it does not accurately describe the bow to define it by its function, since a gun also performs the function, so the scope is not reasonably certain and the claim is indefinite. Similarly, since the specification cannot and therefore does not preclude the existence of the gun, the claim lacks written description, because applicant does not prove he possesses the full scope of the claim. For a similar reason, the claim is only partially enabled and lacks enablement over the full scope. And finally, because it claims the abstract result unbounded by the concrete means of producing it, a claim on the function impedes the art because it chills research into better machinery which achieves the same function (e.g. the gun). And yet I all-too-often see specifications where I’m pretty sure there was a bow in there at some point, but the bow disclosure has been sidelined or entirely done away with because the attorney thinks they can use fancy wordplay to expand the scope against a legal backdrop that explicitly tells them that patentability does not turn on the scrivener’s art. No lawyer can turn an inventive molehill (assuming it is such) into a mountain. You’d be wise to respect the molehill for what it is, rather than trying to fabricate some legendary creation that has no support other that witty wordplay.

    Of course, that’s the big problem a lot of prosecutors have – they either think you CAN invent a function, or they lack the gumption to tell clients that they cannot. (I’m frankly uncertain how much of it is the tail wagging the dog with the client saying “This is the scope we need to clear out our competitors” but that ought to be irrelevant.) Either way, if your client only comes to you with a bow and you use functional language to write a scope that encompasses a possible future gun, you haven’t made your client the inventor of the gun, you’ve just invalidated any would-be protection on the bow. There is no amount of legal skill that turns the bow inventor into a gun inventor.

    The only time a functional novelty would be valid would be in the event that a function is produced, there’s no structure to accurately describe it, and one is certain that there is no other way to cause the function to occur, which simply lacks credulity that it would ever occur. The only time I’m aware of being unable to describe a structure in all of science deals with the heisenberg uncertainty principle, and I doubt that even inventions which utilize that would be otherwise undescribable.

    1. 1.1

      Good stuff, Random, but it reminds me of the jurisprudence under the EPC on the issue of the “problem invention” where the inventive step lies in perceiving the problem to be solved. Such an invention can be patentable even when the solution to that problem, once the problem is addressed, is trivial. When such an invention is patented, it is effectively covering all solutions to the problem. There’s the excitement.

      Only once in a blue moon though, under the EPO’s established case law., such a problem invention is patentable.

      Here is a Link to the Simethicone case from 1983 (soon after the EPO opened for business in 1978) . Simethicone is still the Daddy of all EPC case law in this area. Scroll down to para 6, et seq, in the “Reasons” section of the Decision, for recognition of the patentability of the added “barrier means”.

      link to

    2. 1.2

      Your “go-to” example ever remains the overplayed display of your lack of understanding of the Ladders of Abstraction that DOES play out in any claim of value (read that as any claim that goes even one step beyond an explicit picture claim).

Comments are closed.