Cases where Written Description is Satisfied, but Enablement is Lacking

In Ariad, an en banc Federal Circuit is set to determine whether (and how) the written description requirement is separate and distinct from enablement. The questions presented read as follows:

1. Whether 35 U.S.C. ยง 112, paragraph 1, contains a written-description requirement separate from the enablement requirement.

2. If a separate written-description requirement is set forth in the statute, what is the scope and purpose of that requirement?

I’m interested in finding cases where the written description requirement has been critical to the outcome. More particularly, I’m interested in seeing cases where the invention is sufficiently described but is not enabled. The recent major written description cases are seemingly only tied to written description doctrine because the panels chose to focus on that argument rather than enablement. A case-in-point is Ariad v. Eli Lilly. As in most written description cases, Ariad’s patent was also challenged on enablement grounds. However, rather than reaching the enablement question, the panel majority held that question moot in the wake of their finding that the asserted claims were invalid on written description. The same result occurred in University Of Rochester v. G.D. Searle & Co., Inc., 358 F.3d 916 (Fed. Cir. 2004) (holding that “in view of our affirmance of the district court’s decisions [invalidating the claims] on the written description ground, we consider the enablement issue to be moot and will not discuss it further.”). The LizardTech decision held the asserted claim invalid under both enablement and written description doctrines.

Using the language lf LizardTech, what are examples of cases where the specification provides enough disclosure “to convince a person of skill in the art that the inventor possessed the invention” but not enough to “enable such a person to make and use the invention without undue experimentation.”

42 thoughts on “Cases where Written Description is Satisfied, but Enablement is Lacking

  1. 42

    To MM: the text in your posting that caught my eye and prompted my “conundrum” remark was:

    “The most interesting (and correct, IMHO) cases are those which stand for the proposition that even if a prior art reference teaches that Method X did not achieve the desired results (e.g., treatment of cancer), it still anticipates a claim to Method X.”

    I can’t recall any such interesting case. The doctrine of “teaching away” is less potent, in Europe.

    You can get a European “use X” claim when you teach for the first time that substance Y solves objective technical problem Z. It is thougfht that this teaching is a useful enough contribution to the art to justify the grant of patent rights.

    But that’s when there is no hint or suggestion in the state of the art, to consider using Y to solve Z. What should happen if the state of the art reports “We tried Y to solve Z but we failed” I don’t know. That’s why I wrote “conundrum. MM do you have particular cases in mind?

  2. 41

    MM: Yes, I think you’re right (Although the caselaw on “use claims” is still quite thin). I guess I’m missing your point. Sorry.

    MD, you asked a question and I answered it. Then you inferred that I was “skeptical” of “use” claims and I’m not, really, and I was explaining why I’m not skeptical. Why did you think I was skeptical?

  3. 40

    C’mon, Max. You’re a big boy. You know that (1) we’re talking about law, not logic; and (2) life is not fair.

    But seriously, from the public’s perspective, why grant Inventor A yet another patent for something that would have necessarily become obvious just months after Inventor A’s filing, with the publication of Inventor B’s earlier invention? You have to draw the line somewhere; although I take full account of Europe’s approach in discussing filing strategy with my clients, I’m not convinced that the EPC’s approach is any more logically sound.

  4. 39

    MM: Yes, I think you’re right (Although the caselaw on “use claims” is still quite thin). I guess I’m missing your point. Sorry.

    Big: My point is that Europe’s 102 and 103 provisions perform complementary and mutually exclusive roles. In Europe, 102(e) art is treated the same, whether it is from the same or different inventor, same or different person, same or different Applicant. But you already knew that didn’t you.

    For the PHOSITA, at the date of the claim, how in logic can their invention be rendered obvious by teachings that have still to be made available to any member of the public? On this point, US law seems to me to be just, well, unfair to inventors.

  5. 37

    “If it’s obvious, it ain’t patentable. Obviousness in Europe is judged relative to the entire prior art universe, taken as a whole.”

    Fascinating.

    In the US, if it’s obvious, it ain’t patentable. Obviousness in the United States is judged relative to the entire prior art universe, taken as a whole. Including 102(e) art. Unless that’s owned by the same person.

    But I think I must have missed your point.

  6. 36

    MD Malcolm, I discern that you are sceptical of the shamelessly exploited European “use” claims. The grand-daddy of these is the EXXON “friction-reducing additive” case in which Exxon found out that (surprisingly!!!) a corrosion-inhibiting fuel additive was also actively reducing friction inside the engine.

    Aren’t these treated essentially as method claims for litigation purposes? i.e., if there is a previously disclosed composition without a known non-trivial use and one finds a non-trivial industrial use for it, then that does represent innovation. But the reward for the innovation is not rights to the composition per se. If I use the composition for something other than the use recited in the claim, surely I do not infringe …

  7. 35

    Excellent conundrum from Mooney, but doesn’t convince me that we’ve got it wrong in Europe, or that different enablement standards are necessary to keep the system healthy.

    If it’s obvious, it ain’t patentable. Obviousness in Europe is judged relative to the entire prior art universe, taken as a whole.

    So, why do we need a novelty provision at all? Answer, to mediate between rival filers, all of whom have filed, but none of whom have yet published anything.

    Malcolm, I discern that you are sceptical of the shamelessly exploited European “use” claims. The grand-daddy of these is the EXXON “friction-reducing additive” case in which Exxon found out that (surprisingly!!!) a corrosion-inhibiting fuel additive was also actively reducing friction inside the engine. That insight was thought worth a patent , but exxon had to go all the way to the EPO Enlarged Board of Appeal to get one.

    And see the UK Supreme Court Decision on the Angiotech patent for a TAXOL-coated stent. That too was found by the High Court and the Court of Appeal to be obvious. It just scraped through the Supreme Court, to the dismay of many. But thus it shows the limits in Europe of what’s patentable over prior art, when that art was already pointing in the right direction and the inventor did no more enabling than to make “plausible” the meritorious “solution to the problem”.

    The law on enablement in Europe is still being refined. We’ve only been at it since 1978. It’s a work in progress, and the climate for corporate filers at the EPO is getting colder all the time.

  8. 34

    MD Are they different (forgive my ignorance) and, if so, how, and why?

    They are different. The standard for “enablement” of an anticipatory reference is lower than the standard for enablement of a claim.

    How much lower? I would say “not much lower.” The most interesting (and correct, IMHO) cases are those which stand for the proposition that even if a prior art reference teaches that Method X did not achieve the desired results (e.g., treatment of cancer), it still anticipates a claim to Method X.

    As for “why” the answer is that granting a patent confers a powerful right on the patent owner. It’s already quite simple to overcome most anticipatory rejections. Why make it even simpler to remove a method or composition from the public domain? Is it really “inventive” to show that a previous researcher couldn’t analyze and perform his own experiments correctly? Should one be entitled to own a prior art composition merely because one was the first to discover a use for it? Our legal system has wisely decided that the answer is “no.”

  9. 33

    MaxDrei, an excellent question. I must admit, I do not know the answer. However, the enablement standard for prior art is unconfined by a statute, while the Section 112, p. 1 enablement is. This could result in differences, such as a fully enabled disclosure in a patent could be insufficient because there was an inadequate written description of how to make and use even though those of ordinary skill, applying available knowledge, would know how to make and use. In such a case, the disclosure would be good for prior art purposes, but could not support a patent.

    On its face though, the result is perverse — which is why I think real question in Ariad should be whether what was described was enabling. Otherwise, the written description requirement (of how to make and use) is without a measurable standard.

    The written description of the invention part of Section 112, p. 1, serves a different purpose: to identify the invention and to distinguish it from the prior art.

  10. 32

    Readers, a plea for enlightenment.

    On the “Matrix” thread I am in a clinch with Ron Katznelson about whether FtF destroys innovation. I got talking about satisfying the enablement requirement in Europe. This started me wondering about the differences there are, in the USA, between sufficiency and enablement.

    If I understand it right, a prior disclosure can’t deprive a claim of novelty, if the prior disclosure doesn’t enable what it discloses.

    likewise, an app that doesn’t enable what it claims isn’t enough to satisfy the enablement requirement.

    But these two standards of enablement are different (if I understand it aright).

    Are they different (forgive my ignorance) and, if so, how, and why?

  11. 30

    “Regardless, the biggest question I have with this line of argument by the critics of Ariad’s invention is why should it apply to biotech inventions and not to mechanical inventions?”

    Bob,

    I don’t believe the Federal Circuit is intentionally applying a different standard for “written description” to biotech v. mechnical. (That may be happening unintentionally.) Instead, what we’re likely seeing here is a difference in “predictability” of what the invention is as described. For mechanical inventions, it’s hard to imagine the structures not being predictable as to what they are as described. In the biotech area, simply describing “humnan DNA that codes for insulin” (I’m using the University of California v. Eli Lilly case as my example) doesn’t necessarily tell you what it’s chemical composition (nucleotide sequence) is. In other words, simply describing “human DNA that codes for insulin” doesn’t tell you sufficiently (adequately) what the molecule is chemically.

    In my view, one of the really troublesome decisions by the Federal Circuit in the biotech/”written description” area is University of Rochester v. G.D. Searle. I have no problem with holding the patent claims in that case to the method for treating pain invalid under the first paragraph of 35 USC 112. My problem with this case is that it should have been based on lack of “enablement” instead of lack “written description.” The fact is the method was adequately described (compounds which inhibit COX-2 but not COX-1). But the patent, at best, only provided one example of a compound that would inhibit COX-2 but not COX-1, and provided no way to screen for such activity except by “trial and error.” That’s O’Reilly v. Morse all over again which, as Michael Risch has correctly observed, is an “enablement” case, not a “written descripton” case.

  12. 29

    Sean–

    There’s the rub. In lack of enablement cases, sufficient WD always has to be inferred, as it is never expressly found to exist.

    I don’t believe that such an inference is ever appropriate, considering how courts actually decided cases. For purposes of precedent, one cannot rightly infer an outcome in an issue on which the court was silent.

  13. 28

    “In re ‘318 Patent Infringement Litigation, don’t you think?”

    Kevin,

    Probably. The Federal Circuit sort of skipped over whether the “written description” requirement was satisfied. They also kind of conflated the utility requirement under 35 USC 101 with the “how-to-use” language in the first paragraph of 35 USC 112. But the Federal Circuit did say they were affirming the district court based on lack of enablement which suggest primary reliance on the “how to use” provision of the first paragraph of 35 USC 112.

  14. 27

    PGS v. DeKalb, where the CAFC punished the patentee for not enabling the “full scope” of the claimed invention. In the Court’s view, essentially the inventor had provided an enabled species, but claimed a genus for which another species was not enabled. Although there is no explicit discussion of WD, it can be inferred that the WD was sufficient in this case.

    link to ll.georgetown.edu

  15. 26

    Bob: That is, the critics argue that granting patent protection to Ariad’s invention will stifle research and development of the concept. But this can always be argued about any patent

    Yes, it can always be argued but the argument is compelling in cases like Ariad’s where the claims are broad and the support for the claims is scant.

  16. 25

    I think there was a discussion of this in the district court decision in Amgen v Hoechst Marion Roussel (D Mass), but I don’t remember how it came out.

  17. 24

    Re the Morse case and the comments by EG, Michael Risch, and Ned Heller:

    Current written description rejections and some cases are finding a lack of written description, but not a lack of enablement, where all of the means for practicing a broad invention are not clearly or explicitly describe in the application. That is, the patent office and the Federal Circuit are requiring all of the means to be described. This is true even in cases where those of skill in the art could arrive at all of the means after the fact without the need for undue experimentation.

    I do not know if the various means for practicing Morse’s invention would or would not have required undue experimentation, but for many mechanical inventions all of the various different configurations and dimensions that could accomplish the inventor’s purpose based on a single example in the specification would almost always be enabled. The separate written description requirement as applied in the Ariad case (and in the biotech context in general) is qualitatively the same as saying that the mechanical invention lacks written description because the specification does not describe all of the possible dimensions and configurations of the invention.

    The reason for this distinction in application of the written description requirement is often given as the difference in the level of unpredictability for mechanical inventions versus biotech inventions (and more specifically, the difference in the ability to predict structure from function). But if this is the crux of the distinction, then I say we need standards to judge the “predictability” factor. Current caselaw on written description fails to have any analytical standard. And the Federal Circuit’s second question in Ariad does not seem to have been given useful attention in the en banc case (I will keep my fingers crossed that a fair and useful analytical approach will emerge in the Ariad decision, but I am doubtful).

    In any case, I believe that the predictability issue is not the most useful or fair basis for analyzing written description. If a mechanical inventor and a biotech inventor both provide a valuable concept of an invention and both provide enablement for the full scope of their inventions, I say the public has gotten what they need in exchange for the patent monopoly granted to the inventors. Most of the contrary views that inventors like my biotech inventor and the inventors in Ariad do not deserve or should not get patent protection for their broader inventive concept seem to me to be policy based. That is, the critics argue that granting patent protection to Ariad’s invention will stifle research and development of the concept. But this can always be argued about any patent: the public is always not free to practice an invention and thus the public is limited (unless they compensate the inventor and take a license). This is an eternal issue about patent policy and is not limited to written description.

    Regardless, the biggest question I have with this line of argument by the critics of Ariad’s invention is why should it apply to biotech inventions and not to mechanical inventions? What is the distinction in scope of claims, (supposed) chilling effect on development of the inventive concept, and fair contribution to the public by the inventor between my mechanical inventor and my biotech inventor? There is none. The arguments amount to anti-biotech policy arguments.

  18. 23

    Oh, and if you meant the sequence you put, then there are lots of cases. Wands would be an example since the assay was clearly described–there were just no starting materials for it.

  19. 22

    Dennis, I assume you mean the opposite, where the invention is enabled but not described (that is the case where WD doctrine is outcome-determinative). How about Gentry Gallery?

  20. 21

    In cases where patents by independent inventors are lacking enablement, patents should be awarded to the actual inventor for a development period of twenty years then if an inventor and a development team are successful, a 14 year patent should issue. In the event that no developers are interested in a development agreement with the inventor, rights can be auctioned at any time during the twenty years after issuance. The prime time is years after issuance if the invention is good. As the years tick by in the twenty years, the likely hood of development diminishes therefore the minimum percentage of profits should accordingly be reduced to spur agreement. After the twentieth year, the invention would be free to any developers. This system is vital to maintain inventor interest in the financial profitability incentive to produce invention thereby advancing all fields of human endeavor.

  21. 20

    “Contrary to those who argue that Morse is a description case, I think that case shows a described patent that is not enabled.”

    Michael,

    Very astute and right on point. I was frankly surprised to see O’Reilly v Morse characterized in the briefs and oral argument in Ariad as a “written description” case.

  22. 19

    Dennis,

    Although focusing on a separate 112 issue, I believe that a recent article in the AIPLA Quarterly Journal by Joseph Root regarding Best Mode Compliance may contain some examples (Ducking the Asteroid: Practical Steps Toward Best Mode Compliance, Vol 36, No 4, Fall 2008.

    Hope that helps.

  23. 18

    Howabout this: in general, premature filing chemists are the ones who will encounter “enablement” issues. Art-ignorant engineers, who claim one invention on filing, then switch to another one during prosecution, one they weren’t thinking of when they wrote the app, are the ones who will face “sufficiency” problems.

  24. 17

    Delicious Dennis. Many thanks. I had been thinking that “sufficient description” and “enabling disclosure” are two ways to say the same thing and that “Written Description” is something different, namely, whether the app as filed demonstrates that the inventor had already on the filing date specified in the app, with proper precision, the invention which the app seeks to protect.

    But now you write that you want cases:

    “…where the invention is sufficiently described but is not enabled.”

    thereby conflating “sufficiency” and “enablement”. Which leaves me wondering just exactly what is the difference between sufficiency and enablement.

    So, I’m going to enjoy the developing thread all the more. Thanks again.

  25. 16

    ICU Medical, Inc. v. Alaris Medical Systems, Inc. (Fed. Cir. 2009). “Spikeless” connector claims nuked by CAFC panel based on lack of written description. Although the panel had much to say about LizardTech, the holding based on lack of written description instead of non-enablement may have been because puncturing a plastic IV bag with a spikeless connector does not require much “enablement”.

  26. 14

    This situation is one that divides inventorship from invention development.Issue here is what is fair to both. The other issue is what obligation the developers have to include the inventor in any percieved lacking of enablement to complete the invention. Another is are the developers trying to cut the inventor out of the deal and steal the invention Witch I suspect is the normal situation.The conclusion should be rather clear for judges or legislators the developers should not engage in development without cooperation and written agreement with the inventor.

  27. 12

    Dennis,

    One good example is a method-of-treatment claim that lacks supporting in vivo data in the specification.

    “I claim: A method of treating hypertriglyceridemia comprising administering an effective amount of compound X to a mammal in need thereof.”

    If compound X is described, this claim would satisfy the written description requirement. On the other hand, it may fail the enablement requirement unless the applicant can provide data to show that the compound can actually lower triglycerides in a mammal (or has some other in vivo or in vitro effect that correlates to lowering of trigylcerides in mammals).

  28. 11

    Isn’t the Ariad case about the enablement?

    Reducing NF-ฮบB was the invention. It was described. It was claimed.

    What was not described was how NF-ฮบB was to be reduced. As I understand the case, if one of ordinary skill in the art would know how to reduce NF-ฮบB (without undue experimentation) given what was disclosed, the patent would not have been held invalid.

    Let’s assume for the sake of argument that it was well known from the published literature how to reduce NF-ฮบB. I assume the result would have been entirely different. This illustrates that the central problem in Ariad was lack of enablement that stemmed from a lack of written description of how to make and use WHEN a written description was required to enable one to make and use in point of actual fact.

    If the inventors in Ariad knew how to reduce NF-kB, but fail to disclose it, we also have a lurking “best mode” issue even if Ariad could show that those of ordinary skill in the art at the time could reduce NF-kB without undue experimentation:

    They provided not one single example.

  29. 10

    Contrary to those who argue that Morse is a description case, I think that case shows a described patent that is not enabled. Morse clearly described an invention in hand that transmitted printed communication by electromagnetic signal. He had the possession in hand, and had clear conception of it. However, he did not enable the broad scope of ALL such communications, but just the one version he described.

    Starting materials cases might go the same way, you can describe the invention without providing the starting materials necessary for enablement (e.g. anti-Wands).

  30. 9

    It’s dicta, but there’s a helpful discussion of this in In re DiLeone
    58 C.C.P.A. 925, 436 F.2d 1404 Cust. & Pat.App. 1971. See footnote 1 of the opinion.

  31. 8

    Would not such a desciption without an adequate enablement be the case for almost any chemical application with a fully descriptive and original generic claim but lacking what an examiner or judge considers a sufficient number of species examples to support it?

  32. 7

    112–

    A couple of years back I received an OA that did not directly mention enablement, but in which a 112 rejection was made.

    The 112 rejection was for not precisely describing the metes and bounds of the invention.

    I could see many cases where, although (according to an examiner) the precise scope of the claims is not ascertainable due to a vague written description, the enablement requirement could be met–maybe something where the claim involves a means for, or an open-ended temperature (i.e. above 200C), or something similar.

    Having said that, the particular metes and bounds rejection that we received was without merit–but I could envision scenarios where enablement was satisfied but 112 was not.

    I think that both breeds are probably encountered regularly during prosecution.

  33. 6

    Dennis, define enablement.

    Section 112 paragraph 1 requires a written description of two things: 1) the invention, and 2) how to make and use the invention.

    Is your question concerning enablement directed to whether the written description of how to make and use the invention adequate? How could it be adequate if one of ordinary skill could not make and use the invention?

  34. 5

    The cases I’m thinking of are the litigated ones where a broad generic claim has suitable written description support but isn’t enabled for the later-developed competitor product that the patentee wants its claims to cover. e.g., a claim to “An antibody that binds to protein X” with plenty of description and experimental support for such an antibody, except the patentee asserts it against a humanized version of the antibody that nobody could reliably make as of the patent’s priority date.

  35. 4

    Concur with 112 above, isn’t that exactly what we had in Regents v. Eli Lilly that started this mess?

    Any application that discloses a complex molecule but lacks any description of making it will probably meet this criteria – satisfy WD but fail enablement.

  36. 3

    It is 112, pretty much any simplistic originally claimed mechanical invention that has enablement problems is an example for D. You aren’t likely to draw a WD rejection on some simple assembly of parts although you may not come up with something that actually works.

  37. 2

    “More particularly, I’m interested in seeing cases where the invention is sufficiently described but is not enabled.” Isn’t the opposite case thought to be the rarer breed?

  38. 1

    Actually, as I understand it Ariad no longer disputes 1 in its reply brief. It is no longer an issue. The only issue that remains is whether or not Ariad’s circular argument stating that “there is sufficient written description if enablement is present, but there is a seperate written description requirement although it just so happens that it is satisfied when the claim is enabled.” will be called out by certain justices in addition to all the rational judges on the panel that have already practically committed to a position.

    patentdocs has all the info on the case and a copy of the reply brief wherein Araid says “our bad, our words we used are a bit misleading, what we really meant was something much more confusing than you originally thought, but at least now it is somewhat believable”. Give it a read, it is worth the laughs.

Comments are closed.