CAFC: Prior Art Requires Written Support for Jepson Claim

“The CAFC pointing out that you cannot get a Jepson claim on things that are purely fictional, like a time machine, is hardly insightful and comes across as entirely ridiculous, desperate and at least a little bit insulting.”

Stressed businessmanToday, the U.S. Court of Appeals for the Federal Circuit issued a precedential decision in In re: Xencor, Inc. affirming the Appeals Review Panel (ARP) of the Patent Trial and Appeal Board (PTAB), which rejected of claims from a Xencor patent application as unpatentable. This descriptive characterization of what transpired is not particularly shocking—it is hardly breaking news for the Federal Circuit to affirm the PTAB or the ARP, thereby rendering patent claims dead. But if you dig a little deeper at least one aspect of the ruling is nearly breathtakingly appalling. The Federal Circuit agreed with the ARP that Xencor’s Jepson claim preamble lacked written description support in the specification, which rendered the claim unpatentable.

Jepson claims are famously limiting and are typically disfavored in the United States among most clients and practitioners because they are so limiting. Of course, the reason that a Jepson claim is limiting is because everything in the preamble is admitted as being in the prior art, with the claimed innovation being limited to what comes after the transition, which is generally some form of express admission in the form of: “wherein the improvement comprises.” Thus, it is easy to understand why the preamble of a Jepson claim is limiting and admitting what comes before is in the prior art—because the applicant is admitting that only what comes after the transition is an improvement.

Effective today, the Federal Circuit says that you must disclose and describe the prior art, or your Jepson claim is invalid. Obviously, this means that pretty much every Jepson claim in existence is now invalid. It also ensures that Jepson claims will never be used again. If Jepson claims are both extremely limiting and extremely fragile, they simply cannot be used ever again, period.

It is also a mystery why the Patent Office would ever want a decision like this. I suppose it is understandable as to why the Solicitor’s Office defended the ARP, which includes the Director of the USPTO, the Commissioner for Patents, and the Chief Judge of the PTAB, but advocating in favor of requiring the prior art to be disclosed, discussed and supported in a Jepson claim is precisely contrary to the intent of the rules themselves. As many practitioners no doubt know, the Office prefers Jepson format and Rule 1.75(e) specifically encourages Jepson claim format, going so far as to say whenever possible, “as in the case of an improvement”, an independent claim should be written in Jepson format.  Why the Office has previously encouraged Jepson claims is easy to understand—they are easier for the examiner because they specifically and precisely identify what the applicant believes the improvement is and limits examination to only what the applicant specifically admits is the only thing to distinguish over the prior art. What the ARP was thinking is harder to reconcile with thoughtful examination policy.

But it gets worse. To justify this lunacy the Federal Circuit actually pointed out that a patentee cannot obtain a Jepson claim with a preamble that includes a time machine without describing a time machine! Of course you can’t get a Jepson claim on a time machine! You also can’t get a Jepson claim on a unicorn, teleportation device (i.e., the Star Trek transporter), or a method of traveling faster than the speed of light. The CAFC pointing out that you cannot get a Jepson claim on things that are purely fictional, like a time machine, is hardly insightful and comes across as entirely ridiculous, desperate and at least a little bit insulting. You can’t get any kind of claim on a time machine, unicorn, transporter or method of traveling faster than light because they do not exist.

How the Jepson Claim Debacle Unfolded

In February 2020 Xencor filed a patent application with the U.S. Patent and Trademark Office claiming priority to February 2008 and claiming methods of treating a patient by administering an anti-C5 antibody, which binds human C5 protein and provides anti-inflammatory effect that has been researched for its potential use in treating asthma and transplantation. The specification of the filed patent application discloses that antibodies for amino acids, which require multiple injections over long time periods, are frequently involved in the treatment of autoimmune diseases because they can be administered over fewer treatments.

Having disputed patent examiner conclusions on whether certain claim preambles were limiting through the entire ex parte appeals process, Xencor received a decision from ARP in May 2024 finding that claims 8 and 9 of the patent application lacked sufficient written description support in the specification. Claim 8, which includes the Jepson claim preamble that recites “treating a patient by administering an anti-C5 antibody with an Fc domain”, was not supported because the specification did not provide a representative number of species of the claimed genus of anti-C5 antibodies. This same reasoning was applied to claim 9, which includes a similar preamble without the familiar Jepson transition— “the improvement comprising”.

Although the Federal Circuit’s analysis began with claim 9, it’s the appellate court’s ruling on the written description requirement for Jepson claim 8 that will likely create the most controversy. Xencor argued to the Federal Circuit that the written description requirement of 35 U.S.C. § 112 only requires the specification to support the invention as described by the claim language following “the improvement comprising”, which makes all the sense in the world because the exclusive rights provide by a Jepson claim are limited to the specifically identified improvement, not to the underlying and admitted prior art. Unfortunately, agreeing with the ARP that Jepson claim preambles require written description support, the Federal Circuit held that a Jepson claim form limits the improvement by the subject matter of the preamble, which makes the entire invention the claimed improvement as applied to the prior art.

“If we were to accept Xencor’s argument, we would be doing away with the requirement that the written description ‘clearly allow persons of ordinary skill in the art to recognize that [the inventor] invented what is claimed…’ If an inventor were permitted to simply assert, without showing, that she possessed what is claimed to be in the prior art, she might be able, improperly, to obtain a patent on something she does not actually possess.”

The Federal Circuit further agreed with the ARP that Xencor had not established that the anti-C5 antibodies in the Jepson claim preamble were well-known in the art, finding substantial evidence backing the ARP’s ruling in this regard. Xencor’s position that it did not need to provide written description support for what already existed in the prior art, although perfectly logical and entirely reasonable, was simply not persuasive in the view of the Federal Circuit.

Game Over, Time to Go Home

On the one hand the Federal Circuit has really jumped the shark with this ruling. On some other level, however, this ruling is perfecting in line with where the Federal Circuit is trending. No patent is safe, every claim is vulnerable, and why won’t you people just get the message that we—the Federal Circuit—just don’t like patents anymore.

God help us if we need to start providing written support for the prior art or risk invalidity or unpatentability. We might as well just close shop and go home because the game is over.

 

Image courtesy of DepositPhotos.com
Author: kieferpix
ID 294601918

Share

Warning & Disclaimer: The pages, articles and comments on IPWatchdog.com do not constitute legal advice, nor do they create any attorney-client relationship. The articles published express the personal opinion and views of the author as of the time of publication and should not be attributed to the author’s employer, clients or the sponsors of IPWatchdog.com.

Join the Discussion

16 comments so far.

  • [Avatar for TFCFM]
    TFCFM
    March 18, 2025 12:30 pm

    It seems to me that it doesn’t matter a whit whether the Jepson claim for

    “In a method of treating a patient by administering an anti-C5 antibody with an Fc domain, the improvement comprising said Fc domain comprising [the disclosed modification]”

    is objected-to on “written description” or “enablement” grounds. The claim satisfies neither the written description requirement or the enablement grounds for the simple reason that neither the specification nor the prior art (or even the combination of the two) credibly describes any “treatment” for which the modified antibody is useful.

    Sure, the specification guesses at some broad categories of diseases for which the modified antibody might, conceivably, someday prove useful (or, at least, has not yet been shown NOT to be useful), but this is not evidence that a useful treatment is either “enabled” or “possessed” by the applicant.

    (In fact, this is an old trick that, I’m almost embarrassed to admit, I and colleagues attempted to use many decades ago. A pharma company client was “discovering” previously-unknown human genes, and we claimed “treatments” involving administration of various possible-to-make-once-you-know-the-gene-sequence molecules {antibodies, peptides, iRNA molecules…} for broadly-worded classes of disorders — precisely with the idea that if our client {or a competitor} someday discovered that one of these molecules ACTUALLY WORKS for a disorder in one of the broad classes, then we’d at least have a shot at claiming that “treatment” in a still-to-be-pending continuation application belonging to our client. The lengthy specification contained tedious lists of gazillions-of-potential-molecules and nearly-every-disorder-known-to-man to “keep our client’s options open” and to claim {literally} written description support. Of course, at the time we filed the original “here’s-a-gene-sequence-we-just-discovered” application, our client neither “possessed” any later-found-real treatment nor could it tell a skilled artisan how to practice such a treatment. If such a strategy was ever credible, it certainly isn’t today — nor should it be.)

  • [Avatar for Anon]
    Anon
    March 18, 2025 08:23 am

    To all:

    Awhile back, that other blog (does it even exist anymore?) ran a chart showing the use of the Jepson claim format is US practice.

    As I recall, the use was asymptomatically approaching zero.

    Is this decision – and make no mistake (despite those being confused as to why), it is a wrong decision – worth the print? Should anyone even be using this claim format at this point in time?

  • [Avatar for Max Drei]
    Max Drei
    March 17, 2025 04:39 pm

    I’m now confused by the thread. Perhaps somebody can explain how US law prevents an Applicant at the USPTO from amending during prosecution to claim an invention which, even though enabled, was not derivable by a reader of the patent application as filed, that is to say, not disclosed to such a reader, as the contribution to the art, in the application as filed. Under the (first to file) EPC, it is its Art 123(2) prohibition on “adding matter” during prosecution that prevents Applicants going to issue with claims not entitled to the filing date, claims directed to an invention neither described nor defined in the patent application as filed. Perhaps the most striking example of EPO practice is in the area of a claim directed to an “undisclosed intermediate generalisation”. For example, your claim as filed recites “antifolate”, the subject of the sole worked Example in your application as filed is your own product ALIMTA (pemetrexed disodium) and nowhere do you state that the antifolate can be pemetrexed as such. Nevertheless, during prosecution, you amend claim 1’s antifolate to pemetrexed.

    I had supposed that, in the USA, it was the WD requirement that put a brake on such amendments. I had supposed that, just like at the EPO, the WD test is not a question of what is obvious but, rather what is novel relative to the application as filed.

    But now I’m thinking I must have misunderstood. Who will help to clear my confusion?

  • [Avatar for Randall R Rader]
    Randall R Rader
    March 17, 2025 03:02 pm

    Written Description (WD) law strayed from the statutory language years ago in the 2010 ARIAD decision after the 1997 LILLY decision blindly misconstrued both the statute and the case law. WD was never supposed to be a free-standing doctrine different from enablement. The statutory language is clear on that point. Once a court strays from the statute, the absurdities just pile on top of each other. This case illustrates that point perfectly.

  • [Avatar for Stephen Schreiner]
    Stephen Schreiner
    March 17, 2025 11:27 am

    Some of the comments seem to be conflating enablement with written description. I believe this case was decided on written description. See Ariad (Fed. Cir. 2010), Regents of U-Cal. v. Eli Lilly (Fed. Cir. 1997).

    The death knell for these claims was when the Court agreed with the ARP that the preamble language of “treating a patient” was not just limiting, but that it means “treating all patients and all diseases.” That goes too far. The CAFC’s reasoning for that seemed off, too. The Court says that the spec doesn’t say too much regarding particular patients and particular conditions/diseases, and thus the preamble is construed to require written description support for everything. The spec discloses three classes of disease that can be treated using the invention but that is apparently not enough.

    In essence, the Court is finding the claim claims a genus (all types of patients, all types of diseases). Once that occurs, it’s nearly impossible to satisfy written description. This seems common in the unpredictable arts. It wouldn’t happen in the predictable arts. For example, if the claim was to a “method of reprogramming a processor,” a single embodiment would suffice for written description. A court wouldn’t require a showing of all ways to reprogram all types of processors.

    Is it fair to treat life sciences so differently from other technologies? I don’t know. That’s a legitimate debate.

    On Jepson, I agree with Gene that the Court goes too far in saying that there must be full-scope written description support for the preamble. This case will discourage Jepson claiming.

  • [Avatar for Max Drei]
    Max Drei
    March 16, 2025 06:10 pm

    I’m in Europe where the default form of claim is the EPO “2-part” form in which the two parts are separated by the words “characterized in that ” (or “characterized by”). Some people mistakenly suppose therefore that the EPO requires a Jepson, every time.

    I’m following TFCFM’s argument, however, that i) if the specification fails to enable the claimed subject matter then the claim is bad, ii) merely admitting as prior art the combination of technical features set forth in the first part of the claim is not enough to enable that feature combination, and iii) if the feature combination in the claim’s first part is not enabled then neither is the subject matter of the claim as a whole. An admission that the prior art includes a stated combination of technical features isn’t enough to enable that feature combination.

    Far from “asinine”, all that seems to me indisputable, nothing more than common sense, as is TFCFM’s warning about the mischief of allowing such claims:
    “I don’t know what treatment my ‘improved’ antibodies might be useful for, but if somebody else invents a useful treatment that uses them, that method of treatment is MINE!”

    Who will put me and TFCFM straight?

  • [Avatar for K. David Crockett]
    K. David Crockett
    March 15, 2025 12:06 pm

    @Gene Quinn
    I think we are in agreement that doctrinal consistency is too much to ask of the Federal Circuit. If they had decided this on the basis of lack of enablement for the preamble (there are mere hints in the opinion), we would all just be considering a factual issue, and ignoring this case because it would be fact-specific. The case as it is brings into question the validity of many Jepson claims.

  • [Avatar for TFCFM]
    TFCFM
    March 14, 2025 04:52 pm

    It seems to me that “admitting” that something is prior art does not enable that something.

    Let’s try it out: I hereby admit that time machines are prior art.

    (Anyone going to go back and stop me from writing that?)

  • [Avatar for Gene Quinn]
    Gene Quinn
    March 14, 2025 02:28 pm

    @K. David Crockett

    It is one thing for the Federal Circuit (or USPTO) to determine that the claim is invalid because it is not enabled (which would still be questionable in my mind), but it is an entirely different thing to say, as they did, that written support is required for what is admitted as prior art in the Jepson preamble. Requiring written description support for admitted prior art is an over broad ruling with serious consequences.

    Written support is required for what is invented, and with a Jepson claim what is invented (and claimed) is not what is in the preamble– it is what follows the transition. Similarly, the claimed invention needs to be enabled, and with a Jepson claim the claimed invention is an improvement, not what is disclosed for context in the preamble.

  • [Avatar for K. David Crockett]
    K. David Crockett
    March 14, 2025 01:59 pm

    I don’t think it is as bad as it sounds. The preamble in this case was too vague and lacked a written description (and maybe was not enabled). As the ARP stated, the spec did not demonstrate possession of the method mentioned in the preamble. Perhaps a Jepson claim to something like “baking a chocolate cake, the improvement comprising adding cayenne to the dry ingredients” would survive because everybody knows how to bake a chocolate cake?

  • [Avatar for Gene Quinn]
    Gene Quinn
    March 14, 2025 01:34 pm

    @TFCFM writes: “This appears to me to be merely the latest in a long line of cases rejecting attempts to functionally claim a potentially-infinite number of embodiments that an inventor simply has not invented.”

    If only that was what the Federal Circuit said! What they said is you need to have written description support for the admitted prior art, which is asinine.

  • [Avatar for TFCFM]
    TFCFM
    March 14, 2025 11:23 am

    This appears to me to be merely the latest in a long line of cases rejecting attempts to functionally claim a potentially-infinite number of embodiments that an inventor simply has not invented.

    The “improvement” claimed here is including a pair of Fc-region base changes in an antibody that binds a certain protein (a protein named “C5”), and then using that modified antibody in “treatment” of a patient. The inventor apparently recognized that an antibody having the modification “last longer” (crudely simplification) in the patient’s body, which ought to be a good thing.

    Antibodies (again, crudely simplified) are complex molecules that contain at least two regions: a “target”-binding region and an “Fc” region (that has a bunch of functions, none of which are particularly relevant here). The antibodies recited in the claim have the “C5” protein as their target, and the Fc region.

    The Fc is well known and apparently also described sufficiently in the specification, as is the specifically-modified Fc region described in the claim. However, as is well-known, antibodies can have many different (some *very* different) “target-binding” regions, even for a single target. Hence, there are very, very many C5-binding regions that would functionally cause an antibody to bind with C5 protein. Not many are apparently disclosed in the applicant’s specification. This renders this part of the claim functional (the applicant is essentially claiming “any-and-every antibody that has my Fc-region-modification and also has any any C5-binding region) — and this isn’t even the ‘functionality’ to which the PTO objects.

    The applicant goes further and functionally claims “treatment” of a patient using the functionally-defined antibody. The opinion discusses ceased “trials” (apparently using a C5-binding antibody not possessing the Fc-region modification) that failed to demonstrate that a C5-binding antibody could be used to “treat” any condition. It appears to be this functionally-claimed limitation (the recited antibody “treats” anything-or-everything) to which the opinion appears to be directed.

    Claim 9 thus merely-functionally claims a method of (any-and-every) treatment of a patient using any-and-every C5-binding antibody that has the applicant’s-invented Fc-region modification. However, neither the prior art (at least to the extent it is recognized in the opinion) nor the specification (according to the opinion) describes ANY disease or condition that can be “treated” with a C5-binding antibody (Fc-region modified or not). For this reason, it seems fairly inarguable that the subject matter of claim 9 is not adequately described (because nobody — including the applicant — knows what “treatment,” if any, can be effected by the claimed method).

    Simply throwing that unknown and not-yet-invented “treatment” into the preamble of a Jepsen claim doesn’t “fix” this lack of description. To the contrary, doing this HIGHLIGHTS the impropriety of granting such a claim. In effect, both claim 9 (method of “unknown-treatment”) and claim 8 (improvement to “unknown-treatment”) are saying the same thing:

    “I don’t know what treatment my ‘improved’ antibodies might be useful for, but if somebody else invents a useful treatment that uses them, that method of treatment is MINE!”

    The applicant clearly made a potentially-patentable invention to antibodies and should claim the antibodies as a composition of matter. If the applicant can describe a useful and operative method of using its improved antibodies, it should claim that method (“A method of prolonging retention of an anti-C5-antibody in serum of a patient…”) if a non-trivial and operative use for such a method can be identified.

    However, until the applicant (or someone in the prior art) identifies what a C5-binding antibody operatively “treats,” claiming either a “method of treating” or an improvement to such a treatment method using a modified C5-binding antibody remains a time machine — it *might* be possible for someone to invent one in the future, but neither the applicant nor anyone else has yet.

  • [Avatar for Anon]
    Anon
    March 14, 2025 09:40 am

    Yet more patent profanity?

    Add this to the “yes, we really did incorporate everything by reference, but since we did not explicitly REWRITE everything, the Court is going to assert that you really did not include everything and that by not including everything, you intended to change meanings (as opposed to maintaining being broad and highlighting certain aspects within that breadth”

    and you are INVITING applications to be ponderously dense (as well as several hundred pages).

    And of course, THAT will bring about the accusations of “hiding needles in haystacks.”

    Mind you, scriveners (as they always tend to be) WILL likely adjust. Here, I can easily see turning to specialized AI engines that will gorge forth not merely hundreds of pages, but possibly thousands.

    But is this what we really want?

    (hat tip to PeteMoss for nailing it in a legal citation)

  • [Avatar for PeteMoss]
    PeteMoss
    March 14, 2025 09:09 am

    Loom Co. v. Higgins, 105 US 580, 585 (1882) (“When an astronomer reports that a comet is to be seen with the telescope in the constellation of Auriga, in so many degrees of declination, and so many hours and minutes of right ascension, it is all Greek to the unskilled in science; but other astronomers will instantly direct their telescopes to the very point in the heavens where the stranger has made his entrance into our system. They understand the language of their brother scientist. If a mechanical engineer invents an improvement on any of the appendages of a steam engine, such as the valve gear, the condenser, the steam chest, the walking beam, the parallel motion, or what-not, he is not obliged, in order to make himself understood, to describe the engine, nor the particular appendage to which the improvement refers, nor its mode of connection with the principal machine. These are already familiar to others skilled in that kind of machinery. He may begin at the point where his invention begins, and describe what he has made that is new, and what it replaces of the old. That which is common and well known is as if it were written out in the patent and delineated in the drawings.”)

  • [Avatar for Peter Kramer]
    Peter Kramer
    March 14, 2025 05:47 am

    So a PHOSITA who allegedly is familiar with all the prior art needs a description of the prior art in the specification?

  • [Avatar for Pro Say]
    Pro Say
    March 13, 2025 07:48 pm

    Jepson claims: D.O.A.; R.I.P.

    With enemies like the CAFC, who needs Communist China?

Varsity Sponsors

From IPWatchdog