Amgen v. Sanofi: Who Decides Full Scope Enablement

by Dennis Crouch

Amgen Inc. v. Sanofi-Aventis (Fed. Cir. 2021)

Patent claims typically cover an infinite number of potential infringing embodiments.  This seemingly renders true full-scope enablement an impossible task.  But the metaphysics are an illusion.  If we want valid patents, then there has to be some “good enough” threshold for enablement.

The focus in Amgen is a particularly tricky type of claim: genus claim with functional limitations.  Here, the claim is directed to an isolated monoclonal antibody defined by its ability to bind with the protein PCSK9 and consequently block PCSK9 from binding to LDL-C.

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

US8829165 (US8859741 is also at issue). In its decision, the Federal Circuit found the claims too broad in comparison to the disclosure.  “The functional limitations here are broad, the disclosed examples and guidance are narrow.”  Thus, although the jury found the claims enabled, the appellate majority affirmed the district court’s  post-verdict JMOL (notwithstanding the verdict).  This is a biopharma case, but the decision here can also be seen largely as an attack on functional claim limitations across the spectrum of patent technology fields.

Now, Amgen has filed a strong petition for en banc rehearing arguing that the majority has improperly created a new stiffer test for genus claims with functional limitations.

Question 1: Whether the panel’s new enablement test for genus claims with functional limitations, which has no basis in §112’s text, conflicts with Supreme Court decisions, including Minerals Separation, Ltd. v. Hyde, 242 U.S. 261 (1916); Wood v. Underhill, 46 U.S. (5 How.) 1 (1846); and Mowry v. Whitney, 81 U.S. (14 Wall.) 620 (1872), and decisions of this Court and its predecessor, including AK Steel Corp. v. Sollac & Ugine, 344 F.3d 1234 (Fed. Cir. 2003), and In re Angstadt, 537 F.2d 498
(C.C.P.A. 1976).

As mentioned above, the question of enablement was given to the jury.  Here, however, the Federal Circuit held that enablement is a question of law.  The distinction is likely dispositive in this type of close case where the jury sided with the patentee.  By treating enablement as a question of law, the Federal Circuit was able to review the issues de novo without impinging upon the reexamination clause of the 7th Amendment.  The second question challenges this approach:

Question 2: Whether enablement is a question of fact, as the Supreme Court has held, see Battin v. Taggert, 58 U.S. (17 How.) 74 (1854); Wood v. Underhill, 46 U.S. (5 How.) 1 (1846), or a question of law, as this Court holds, Op.6.

[Petition for Rehearing]

The brief cites and discusses a really well researched and written article by Professors Karshtedt, Lemley, and Seymore titled The Death of the Genus Claim (forthcoming 2021).  The article notes that today “it is no longer possible to have a valid genus claim. . . . [This] represents both bad law and bad policy.” Id.  In its decision, the majority seemed to largely agree, but declined to expressly foreclose the possibility of genus claims.

This is a very important case for pharma and biotech patenting, but will also likely have an impact on the doctrine of enablement generally — especially its relationship to functional claim limitations since those are found in most patents.  I expect a number of amicus briefs will be filed in the case over the next few weeks.

More to come.

Functional Claim “Raises the Bar for Enablement”

31 thoughts on “Amgen v. Sanofi: Who Decides Full Scope Enablement

  1. 8

    The key to enablement is trivial to understand to one that wants to understand.

    Scope of enablement and the reverse doctrine of equivalents. Those are the only things needed.

      1. 8.1.1

        Dennis wrote 16 years ago:

        “On appeal, the CAFC paid lip service to the rule that a claim will not be invalidated under Section 112 for only disclosing a single embodiment. However, the court went on to determine that claim 21 was directed at “all seamless DWT’s” — a coverage not adequately supported by the single embodiment.”

  2. 7

    6 different outside counsel (by my count) and a large contingent of Amgen in-house people too. Is that some kind of record?

  3. 6

    While not cited in Amgen’s petition, for a start, Robinson, The Law of Patents, vol. II, Section 503 titled “The Description: its Sufficiency a question of fact”, and n. 1, seems germane to question 2.

    1. 6.1

      Right 19th century folks thought of enablement as a question of fact. That flipped in the 20th century. The question is whether there is any viable explanation.

      1. 6.1.1

        Is there at least a “first case” that reflects the flip?

        I ask because Greg’s comments almost appear to present a type of ‘de facto, well, that’s always been the way’ mindset, and at some point that ‘flip’ had to have a start.

      2. 6.1.2

        “The question is whether there is any viable explanation.”

        Because the CAFC considers themselves to be the kings of everything patents?

        Because, well, you know; kings don’t have to supply viable explanations.

        1. 6.1.2.1

          They have been well trained (as are simians in a cage, a step ladder in the middle of the cage, a bunch of bananas hanging from the top of the cage, above the step ladder, with the Supreme Court Justices brandishing a fire hose, and pummeling any simian who dares to attempt to use the step ladder to reach for the bananas).

      3. 6.1.3

        Amgen cited A.B. Dick v. Barnett (2nd Cir. 1923) from the early 20th Century, which in turn cited Hogg v. Emerson, 52 U.S. (11 How.) 587, 606-07 (1850), for it being a jury question. Merger of law and equity in 1938 should not have changed it. Note Amgen also cited Tights v. Stanley (4th Cir 1971) with its discussion of S. Ct.’s decisions (Dairy Queen, Beacon Theatres, Great Atlantic & Pacific Tea, and Graham).

  4. 5

    Question 2: Whether enablement is a question of fact, as the Supreme Court has held, see Battin v. Taggert, 58 U.S. (17 How.) 74 (1854); Wood v. Underhill, 46 U.S. (5 How.) 1 (1846), or a question of law, as this Court holds, Op.6.

    Rather tendentious of them to contend that either of those cited cases hold that enablement is a question of fact. I do not see that the fact/law distinction is actually at issue in either case.

    Moreover, while Chief Justice Taney cites the enablement language of the 1836 patent act as a preliminary to his analysis of the Wood patent’s sufficiency, the discussion looks a lot more like a discussion of the definiteness requirement. It is not clear to me that Wood is really about the enablement requirement at all.

    I think that all of §112 should be treated consistently (either clarity, enablement, and written description should all be questions of law, or all questions of fact), so I would be delighted to think that the CAFC has been wrong in treating enablement as a question of law. This petition, however, does not make a very convincing case for that point. I looked up those citations expecting to be impressed, and instead I finished disappointed.

  5. 4

    As I have been saying for years, Lemley’s next attack was on 112 trying to remove a person of ordinary skill and common sense.

    1. 4.2

      As I have been saying for years, Lemley’s next attack was on 112 trying to remove a person of ordinary skill and common sense.

      I am not sure what you mean here. Karshtedt & al. appears to me to be arguing for more common sense in the §112 analysis and less of a rigorous adherence to statutory literalism. It looks to me like you are so set in your expectation of what you think Lemley would argue that you are not reading what he is arguing.

        1. 4.2.1.1

          Still I’ll have to look at this. 112 is being used to hurt patents right now and the legislation proposed to fix 101 had a poison of amending 112 to limit claims to almost picture claims.

          Additionally, the CAFC has fabricated a need for a method expressly laid out in the specification when a person of ordinary skill in the art would know the method and understand the differences expressed in the specification.

          Additionally, the CAFC has fabricated a heightened 112 standard for functional claims when ALL persons of ordinary skill work discuss inventions in functional terms.

          Etc.

          So what does a more common sense approach mean? A more common sense approach to me would be looking at how a person of ordinary skill in the art views the specification and not judges at the CAFC who have undermined the patent system with their 112 jurisprudence. They probably unjustly invalided at least 100,000 patents–in effect–with just those two holdings.

          But I am still playing fast and lose and need to get a better grip on the games that Lemley is playing with 112.

          1. 4.2.1.1.1

            [T]he legislation proposed to fix 101 had a poison of amending 112 to limit claims to almost picture claims.

            (1) As they say, one man’s meat is another man’s poison. It is strange to characterize the §112(f) revisions in that bill as the “poison” part of the bill. The §112(f) revisions were “poison” to some, and “meat” to others. Same for the §101 revisions in that bill. If you look at the list of parties who testified in Congress against the bill (e.g., High Tech Inventors Alliance, Association for Accessible Medicines, etc.) it is scarcely likely that it was the §112(f) revisions that were drawing most of their ire. Most of the opposition came from people who opposed the §101 revisions.

            (2) It is not really accurate to say that the proposed §112(f) revisions would have reduced most claims to picture claims. If a competent patent attorney did the work of drafting an application well, the scope of the claims that one would get under the proposed revised §112(f) text would be no less capacious than currently (and more enforceable to boot). The real objection is not that the claims under such a statue would be narrower, but rather that the revised statutory language would expose the difference between prep-&-pros attorneys who know their stuff and those who do not.

            1. 4.2.1.1.1.1

              It is strange to characterize the §112(f) revisions in that bill as the “poison” part of the bill.

              Speak of the devil….

              but rather that the revised statutory language would expose the difference between prep-&-pros attorneys who know their stuff and those who do not.

              Total B$ on that assertion. Greg does not know what the H he is talking about (and is one of those ‘Pharma’ guys that quite frankly would rather separate Pharma from any of that ‘computer’ stuff.

            2. 4.2.1.1.1.2

              Greg >>“but rather that the revised statutory language would expose the difference between prep-&-pros attorneys who know their stuff and those who do not.”

              Total nonsense. The proposed revision would limit a functional claim element to only embodiments disclosed in the specification, which anyone in the EE/CS/Mechanical arts knows means that your claim is worthless as there are too many possible embodiments to disclose them all.

              1. 4.2.1.1.1.2.2

                Total nonsense… [T]here are too many possible embodiments to disclose them all.

                (1) The proposal was that the claim would read on disclosed embodiments and equivalents.

                (2) You control how many embodiments you disclose. Like I said, the real objection is not that the claims would necessarily be narrow, but rather that skill and diligence would be necessary to obtain the full scope to which the inventor is entitled. Too many attorneys did not want to be caught skinny dipping when the tide went out.

                1. Greg, the proposal was to make the scope the same as for means for claims.

                  It is well established law that that scope is much narrower.

                  And you can’t “control” the embodiments you disclose because in the EE/CS/mechanical arts fields there are too many possible embodiments. Those skilled in the art understand this.

                2. Greg’s view on this is but a p00r mirror of Malcolm Mooney’s “only picture claims are ‘real‘ inventions” mantra.

  6. 3

    Amgen says the spec recited 26 different species that taken together represent the diversity of species within the genus.

    If thats true, let’s say they used 112(f) claiming to cover all of the 26 different species AND equivalents of the 26 different species. What does this not get them that a genus claim would?

    Is the real problem that DOE is so underdeveloped?

    1. 3.1

      Should this poor decision stand (expect Amgen to go SCOTUS if it does), it may necessarily result in innovators having to rely on / pray for DoE to block others from infringing.

      Does the CAFC really want to plop such a DoE can-o’-worms to their already full plate?

  7. 2

    “genus” and “species’ are a case of ‘turtles all the way down.

    This is both a problem in the strict “two item” set and the fact that ALL Ladders of Abstraction have more than two rungs.

    1. 2.1

      If you just look at what they are up to anon, this is all about hurting patents.

      The 112 jurisprudence is about taking away how one skilled in the art would interpret the claims.

      1. 2.1.1

        I hear you – and completely agree with your point (here and above in relation to the insertion of 112 changes in the discussion of fixing 101 as very much being a poison pill).

        There was ZERO reason to try to conflate the two items, and that conflation was what sunk and otherwise admirable attempt by Congress to do something.

        Noting exactly who came up with that and why they came up with that easily identifies the culprits who really do not want to fix the problems in patent law.

  8. 1

    This seemingly renders true full-scope enablement an impossible task.

    Otherwise known as “Random’s Fallacy.”

Comments are closed.