Rethinking enablement: Court grants cert in Amgen v. Sanofi

By Jason Rantanen

On Friday, the Supreme Court granted certiorari on two petitions raising intellectual property issues, including the closely-watched enablement Amgen v. Sanofi.  The other case is a Abitron Austria GmbH v. Hetronic International, a trademark dispute involving extraterritoriality issues.

Over the past few years, the Federal Circuit has issued a series of enablement decisions, including Amgen v. Sanofi, that rely on a lack of enablement to hold that functionally defined genus claims are invalid. (Dennis’s post on Amgen is here.This has been a topic to watch at the Federal Circuit, and two major law professor articles, one written by Mark Lemley and Jake Sherkow and the other co-authored by Dmitry Karshtedt, Sean Seymore, and Mark Lemley, have pointed out the problems with the Federal Circuit’s approach to genus claims, especially in the context of inventions such as antibodies. Karshtedt, Seymore and Lemley filed an amicus brief supporting en banc rehearing at the Federal Circuit (PatentlyO post here) as well as one supporting Amgen’s petition for certiorari.

The claim in Amgen illustrates this point. It claims the genus of antibodies that bind to a particular region on a specific protein. Here is a representative claim of Amgen’s Patent No. 8,829,165:

1. An isolated monoclonal antibody, wherein, when bound to PCSK9, the monoclonal antibody binds to at least one of the following residues: S153, I154, P155, R194, D238, A239, I369, S372, D374, C375, T377, C378, F379, V380, or S381 of SEQ ID NO:3, and wherein the monoclonal antibody blocks binding of PCSK9 to LDLR.

The problem is that there isn’t just one antibody that meets this requirement – rather, it encompasses potentially millions of currently unknown antibodies. The Federal Circuit held that the claims were far broader than the associated disclosure, requiring undue experimentation to identify undisclosed embodiments encompassed by the claims. Amgen petitioned the Supreme Court for review, extensively citing the Karshtedt/Seymore/Lemley article.  

On Friday, the Supreme Court granted certiorari on Question 2 of the Amgen petition, which focuses on the “full scope” and “undue experimentation” aspects of the of 112(a) jurisprudence. This was after the Solicitor General had recommended against certiorari.

Here’s the question that the Court granted cert on:

2. Whether enablement is governed by the statutory requirement that the specification teach those skilled in the art to “make and use” the claimed invention, 35 U.S.C. § 112, or whether it must instead enable those skilled in the art “to reach the full scope of claimed embodiments” without undue experimentation—i.e., to cumulatively identify and make all or nearly all embodiments of the invention without substantial “ ‘time and effort,’ ” Pet. App. 14a (emphasis added).

This will be the first time that the Supreme Court has addressed the enablement requirement in a very long time.  Those decisions used the pre-1952 version of the patent law statute, Rev. Stat. § 4888. Generally, however, it was very similar to the current 35 U.S.C. § 112(a). For example, as applied in Consol. Elec. Light Co. v. McKeesport Light Co., 159 U.S. 465 (1895) (a/k/a The Incandescent Lamp Case), it required:

“a written description of the device, and of the manner and process of making constructing, compounding, and using it in such full, clear, concise, and exact terms as to enable any person skilled in the art or science to which it appertains or with which it is most nearly connected to make, construct, compound, and use the same.”

To me, the issue of full scope enablement is one of the most challenging problems in patent law. Many patent law scholars have written about it, including myself, and I still haven’t seen a real solution. But I’m sure that much more will be said before and after the Court issues its decision.

Sadly, as I wrote about last week, Professor Karshedt passed away just a few days ago, and we’ll miss his insights on the case as it moves through the Supreme Court.

Here’s a link to the docket on ScotusBlog for those who would like to read more.

29 thoughts on “Rethinking enablement: Court grants cert in Amgen v. Sanofi

  1. 8

    May I suggest a different perspective on the disclosure requirements and corresponding claims around section 112(a). One that is based on a policy question. Does the science described in the Amgen patents deserve a 15-20 year monopoly to restrict the use, make, or importation of any antibody (not yet made or unknown) that binds to one of the fifteen aa residues (and it does not matter how strongly, or conversely, how weakly) the antibody binds to any one of the listed 155 residues listed in claim 1 above? If your answer is yes to this question, then you believe that the Supreme Court should reverse the Federal Circuit decision. Also, you then must ask yourself what would be a reasonable royalty rate Amgen may extract from others (assuming Amgen is willing to license the patent(s) to others)? Lastly, assuming Amgen is willing to license, and the royalty rate is 15% to 20% of U.S. sale revenues, would these others begin or continue research programs in this area in the hope of developing other like functioning antibodies?

    Its not just the word games we attorney’s love to play or to argue what is “undue experimentation”.

    What I think is a very good analogy in the chemical (pharma) art of drug research, and so much more important than the technology at play here, was the development of statins in the late 80’s through the 90’s. Repatha is at most a $500 million drug with significantly small number of patients. Technically, if you google “statin chemical structures” one immediately sees the number of different statin drugs developed by many different pharma companies, and we all know them: pravastatin, simvastatin, atorvastatin, and rosuvastatin. What one finds in common with all these important drugs is the hydroxy lactone (natural statins) or a corresponding 2,4-(dihydroxy)hexacarboxylic acid group. Moreover, all these compounds are HMG-CoA reductase inhibitors and with different degrees of binding coefficients, and perhaps, more importantly, different half-life or biological metabolites. Now imagine if only one company had U.S patent rights. Do you think Astra Zeneca would have developed a program around rosuvastatin?

    Lastly, where do we stop. In reality, no one really cares about Rapatha, and believe me, neither does Amgen. As I said earlier in a posting, Amgen has similar patents (one may say paper patents) around Regeneron/Sanofi Dupixent, a IL-4/IL-13 antibody that is a life changing for millions and a drug antibody that recently hit 10 billion in annual revenues (likely 8B in the U.S. – that’s just how it is). The Amgen patents that allegedly cover Dupixent also have similar 112(a) issues. Amgen also had an Il-4/Il-13 antibody drug in the clinic for, but it failed to obtain approval. Do these patents pass 112(a) muster? Where does one stop? and why are antibodies so special that they deserve broader patent protection than the life changing statins.

    1. 8.1

      Not following “Also, you then must ask yourself what would be a reasonable royalty rate Amgen may extract from others” in view of your premise that this is a “policy” inquiry.

      What the market would bear is the only “policy” that need apply.

      Perhaps you are not appreciating the “Stick” portion of the patent right.

      The (correct) policy is to pay what the patent holder requests, or go without.

      If the desired trait simply has too high a price tag, then the “keep-out” aspect will promote inventing-around.

      1. 8.1.1

        The issue is how big the stick when your “product antibody” claim has zero (nil) structure to define what you are claiming. In other words, write a structural genus around the commonality of your 20 or so example antibodies that bind to the site of interest just as one would do with a small molecule claim.

        1. 8.1.1.1

          You miss in multiple dimensions vis a vis your reply in regards to the stick of patent law.

          For example, you confuse a discussion point as to whether a claim is valid in the first instance (based on “functional”) with what a valid patent should entail.

          For another example, you STILL do not address the fact that ANY stick aspect is NOT something a court should be meddling with for policy reasons.

          The courts are not where patent policy is to be decided.

          That’s an explicit power distribution made in our Constitution.

  2. 6

    The problem with the claim is that it relies way too heavily on functional language. A valid claim would recite at least some comparison to a reference protein sequence existing in one or more actual antibodies identified and described in the priority document, e.g., “wherein said antibody includes a sequence at least 95% identical to SEQ ID No. 1” or some similar language.

    Note that such a claim is a bona fide genus claim potentially covering many, many species (the “argument” that “genus claims can’t be gotten and enforced” is complete b.s., as should be easily recognized by any attorney in the field).

    Should the Supremes decide (incorrectly) for Amgen, it will be an example of patent law being made much worse for everybody, including consumers who are going to end up footing the bill for the additional (undeserved) patent coverage that all players are going to be forced to navigate around..

      1. 6.1.1

        Apologies for a triple (not double — you copy/pasted the same thing on multiple threads as well) is the least of the things that you should be apologizing for.

          1. 6.1.1.1.1

            Of course you do — not just blame someone else for your issues, but blame the proverbial “everyone else.”

    1. 6.2

      It is where and how an antibody binds to its target that determines its impact, not necessarily the particular amino acid sequence. Once that information is known, design arounds that don’t really add to the art can be fairly easily achieved. And therein lies the rub — the experimentation is not really undue.

  3. 5

    The problem with the claim is that it relies way too heavily on functional language. A valid claim would recite at least some comparison to a reference protein sequence existing in one or more actual antibodies identified and described in the priority document, e.g., “wherein said antibody includes a sequence at least 95% identical to SEQ ID No. 1” or some similar language.

    Note that such a claim is a bona fide genus claim potentially covering many, many species (the “argument” that “genus claims can’t be gotten and enforced” is complete b.s., as should be easily recognized by any attorney in the field).

    Should the Supremes decide (incorrectly) for Amgen, it will be an example of patent law being made much worse for everybody, including consumers who are going to end up footing the bill for the additional (undeserved) patent coverage that all players are going to be forced to navigate around.

    1. 5.1

      So what is that “objective physical structure” of that other 5% (especially as that appears to be the Gist of the “Point of Novelty” that is the aim for the expanded claim scope)…?

  4. 4

    Sadly, it is easy to see how the Court could make enablement as clear and easily understood as they have done for patentable subject matter. Having nine people who know literally nothing about science (especially this science) making decisions that scientists must cope with is an absolutely brilliant idea. Having people who have never done a single scientific experiment in their life (perhaps one of them dissected a frog while in high school) decide what “undue experimentation” means is like having someone who’s declared bankruptcy 9 or 10 times decide fiscal policy. Hold on to your panties everyone because I’ll be shocked (but pleasantly surprised) if this doesn’t get as messy as 101.

    1. 4.1

      Having people who have never done a single scientific experiment in their life (perhaps one of them dissected a frog while in high school) decide what “undue experimentation” means is like having someone who’s declared bankruptcy 9 or 10 times decide fiscal policy.

      Which is why it is strange that the Court declined cert. on question #1 here. Really, why is enablement a question of law? Are history major judges better placed than most jurors to decide how much experimentation is “undue”?

  5. 3

    Describing “how to make and use” a claim should be assessed in light of claim scope. If the claim recites a broad genus, then that is what has to be enabled. I see the question presented as somewhat of a false dichotomoy. Prediction: affirmed under any standard (so if remanded under some new standard, the result will be the same).

    1. 3.1

      Conversely, there is also a canon of claim interpretation that requires the interpretation to preserve the claim validity. It makes sense to interpret a claim in a way that the interpretation meets the written description, enablement, and preferred embodiment requirements of 112(a). A claim can still be found invalid under 112(a) when the Defendent accepts the Patentee interpretation and can then prove it fails 112(a).

      1. 3.1.1

        Re: “a canon of claim interpretation that requires the interpretation to preserve the claim validity.”
        I think you will find modern cases that suggest that canon has shot its wad.

        1. 3.1.1.1

          “While we have acknowledged the maxim that claims should be construed to preserve their validity, we have not applied that principle broadly, and we have certainly not endorsed a regime in which validity analysis is a regular component of claim construction. Instead, we have limited the maxim to cases in which the court concludes, after applying all the available tools of claim construction, that the claim is still ambiguous.” Phillips v. AWH (Fed. Cir. 2005) (en banc) (internal citations removed).

  6. 2

    Nothing more than a coincidence, I suppose, but this very recent item:

    link to ipkitten.blogspot.com

    in a leading UK patent law blog is about enablement in an antibody case, written by an in house at a leading innovative pharma concern and it begins with the words:

    “………a case study of how functional claim language that wouldn’t pass muster in the US, can be used to great effect in Europe.”

    I guess it is problematic for some in the USA, when a US corporation can get commercially powerful claims of wide scope to issue in ROW but not at home in the USA.

    1. 2.1

      Yes MaxDrei, the US Sovereign will treat US patents under US patent law (which will be different than other sovereigns).

    2. 2.2

      Three thoughts:

      1) In the pharma industry, it barely matters what other jurisdictions do. The U.S. is far and away the largest and most profitable market in the world in this industry. If the U.S. patents fail, that is a catastrophe for the innovator, even if the patent holds good in every other jurisdiction.

      2) Do not imagine that this is a difference between the US and ROW. EP is the outlier here. CN/JP/KR do not allow nearly the same breadth of antibody claim scope as EP.

      3) The reason why EP is more accommodating on this dimension than is US has to do with #1. Because of the US’s unique, for-profit approach to healthcare finance, biological drug costs (and profits) are many times those that Europeans pay for the same drugs. This cost gap motivates US judges to distort patent law (uselessly and inefficiently) to try to constrain US drug costs. Because the EP employs a more effective cost control mechanism (direct price caps), the EP courts feel no need to muck around with patent law.

      1. 2.2.1

        Wow! For one who is not within the pharma sector (me) that, sir, is very informative. Many thanks.

        Interesting for me is the interplay of politics with patent law. I had supposed that the EPO ignores the politics, when deciding under the applicable patent statute how much scope the inventor has enabled. Up to now, I had naively supposed that the courts in any given jurisdiction would do that too. Only in Europe though, it seems.

      2. 2.2.2

        EU Gov: “Here’s your broad, gov’t granted monopoly on the claimed subject matter. Congrats! Oh, by the way you can only sell it for $4.”

        US Gov: “Wait a minute, I know you are going to sell that drug for $10k/per treatment. That’s way too expensive. I better revoke your gov’t granted monopoly!”

        An now, any other field other than pharma (US or EU): “Here’s your gov’t granted monopoly. How much are you selling the product for? Oh, never mind.”

  7. 1

    Note that the above Amgen claim does not present the more typical contested enablement case of a genus claim with [what someone considers] too many possible undisclosed species. Here, we have three distinct claim limitations [2 functional] for a “monoclonal antibody” namely:
    1. it is bound to PCSK9, and
    2. binds to at least one of the following residues: S153, I154, P155, R194, D238, A239, I369, S372, D374, C375, T377, C378, F379, V380, or S381 of SEQ ID NO:3, and it
    3. blocks binding of PCSK9 to LDLR.
    [P.S. I thought cert question 1 here, which the Sup. Ct. did not take [the legal vs. factual issue], was of more general interest than this number 2 will be with this claim?]

    1. 1.1

      There are some functional limitations here with no claim structure for how they are met. Many 101 cases go off on just that (abstract idea: block the blocking enzyme by clogging up one of its characteristic surface features). Regardless of the specific merits of this dispute, doesn’t this case demonstrate that this type of question is much better approached under 112 (as it was here)? As for the merits, though, doesn’t this claim leave a practitioner at a loss for how to identify/make any specific compound that actually binds (presumably in vivo) to one of the specified residues?

Comments are closed.