Amgen v. Sanofi: A well characterized antigen insufficient for written description of an antibody

IncompleteIs knowledge of the chemical structure of a novel antigen sufficient to satisfy the written description requirement of claims directed to a genus of antibodies when the antibodies are further defined solely by function (e.g., by their affinity to the antigen)?   Not sufficient, according to the Federal Circuit opinion in Amgen v. Sanofi No. 2017-1480  slip op. Fed. Cir. Oct. 5, 2017 (“Amgen”).  The opinion concluded that this so-called “newly characterized antigen” test essentially dispensed with the written description requirement by equating it to the enablement requirement.  Cognizant that some of its earlier opinions might be viewed as supporting the test, the Court provided context to those opinions and reminded that precedential value of cases in the area of written description was extremely limited, requiring each case involving a written description issue to be decided on its own facts.

An antibody is a large Y-shaped protein that can bind to a small molecule (an antigen) or a small region on a large molecule, (an antigenic determinant) with both high specificity and affinity.  High specificity implies that even a minor change in the antigen structure can lead to significant loss of binding of an antibody to the antigen.  An antibody has two structurally identical arms, each consisting of a so-called light chain and part of a so-called heavy chain. Each arm is capable of binding an antigen.  A portion of each of the light and the heavy chain, the variable region, is capable of enormous variation due to the possibility for different amino acid residues to occupy certain positions in the amino acid sequence of this region.  This leads to an extremely large number of antibodies which differ in structure, and generally also in binding affinity and/or specificity.  Additionally, fixed variations in the non-variable regions of the heavy chain lead to five broad classes of antibodies.

Methods for producing antibodies are routine, which led to the notion that a fully characterized antigen not only describes itself, but also the antibodies that bind to it.  This logic implied that if the antigen is novel and fully characterized, one could claim an antibody to it by just specifying its binding affinity.  Both the USPTO, and to a degree the Federal Circuit, have contributed to this notion.  Amgen at 13, 14.

Turning to the facts of the case, Amgen markets the low-density lipoprotein (LDL) lowering drug Repatha™ which contains a monoclonal antibody called evolocumab as its active ingredient.  Id. at 4.  Repatha binds to the protein PCSK9, preventing it from causing destruction of another protein, LDL receptor (LDL-R).  LDL-R is found on liver cells and functions to extract LDL from the blood stream.  Id.  The patents at issue, U.S. Patent Nos. 8,829,165 (“’165 patent”) and 8,859,741 (“’741 patent”) have claims that cover the entire genus of antibodies that bind to specific amino acid residues on PCSK9 and block it from binding to LDL-R.  Id. at 4, 5.  Notably, the patents do not specifically claim any antibody by its amino acid sequence.  Id. at 5.  Claim 1 of the ’165 patent is representative and reads:

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 blocks binding of PCSK9 to LDL-R.  Id.

The patents describe testing of over 3000 human monoclonal antibodies to confirm stable binding to PCSK9 and narrowing of the number down to 85 to identify antibodies that block interaction between PCSK9 and LDL-R at greater than 90%.  Id.  The patents also disclose the three-dimensional structures of two antibodies known to bind to residues recited in the claims.  Id.  In addition, they disclose the amino acid sequences of twenty two other antibodies that compete with Repatha or another antibody, 31H4, for binding to PCSK9.  Id. at 5, 6.

Sanofi, Aventisub LLC, Regeneron Pharmaceuticals Inc., and Sanofi-Aventis U.S., LLC (collectively, “Sanofi”) developed Praluent, the active ingredient of which, similar to Repatha™, is a monoclonal antibody that targets PCSK9 and prevents it from binding to and causing destruction of LDL-R.  Id. at 6.  Praluent received FDA approval, following which Amgen sued Sanofi for infringement of the ’165 and ’741 patents.  Id.  While stipulating to infringement, Sanofi challenged the patents’ validity on the grounds of written description, enablement, and obviousness.  Id.

The written description requirement issue in this case pertains to the instruction given to the jury by the trial court to determine if the requirement was met.  The jury received a standard instruction stating that to satisfy the requirement the patentee may disclose either a representative number of species falling within the scope of the genus or disclose structural features common to the members of the genus so that a skilled artisan may visualize or recognize the members of the genus.  The jury was further instructed as follows:

In the case of a claim to antibodies, the correlation between structure and function may also be satisfied by the disclosure of a newly characterized antigen by its structure, formula, chemical name, or physical properties if you find that the level of skill and knowledge in the art of antibodies at the time of filing was such that production of antibodies against such an antigen was conventional or routineId. at 12.

Sanofi argued that the instruction was erroneous, contending that disclosing an antigen does not satisfy the written description requirement for a claim to an antibody.  Id.  According to Amgen, the instruction simply restated the law as set forth in the Federal Circuit cases Enzo Biochem, Inc. v. Gen-Probe Inc., 323 F.3d 956 (Fed. Cir. 2002), Noelle v. Lederman, 355 F.3d 1343 (Fed. Cir. 2004), and Centocor Ortho Biotech, Inc. v. Abbott Labs., 636 F.3d 1341 (Fed. Cir. 2011)(“Centocor”).  Id. at 12, 13.  The Federal Circuit determined that the instruction was improper and not based on any binding precedent.  Id.  The Court’s reasoning, both in the earlier cases and in the present case, is explained below.

Enzo Biochem involved claims directed to nucleic acid probes specific to a certain bacteria.  The probes were defined functionally by their ability to selectively hybridize to the chromosomal DNA of the bacteria over that of bacteria of a different species (with a ratio of greater than 5:1).  Enzo Biochem at 961.  Explaining that not all functional descriptions of genetic material fail to meet the written description requirement, the Court cited PTO’s Guidelines for the theory that functional characteristics when coupled with a known or disclosed correlation between function and structure may satisfy the written description requirement.  Id. at 964.  Moreover, in dicta, citing a PTO source, the Court stated that “the PTO would find compliance with 112, [¶] 1, for a claim to an isolated antibody capable of binding to antigen X, notwithstanding the functional definition of the antibody, in light of the well-defined structural characteristics for the five classes of antibody, the functional characteristics of antibody binding, and the fact that the antibody technology is well developed and mature.” Id.

Noelle involved a patent application that included a claim directed to a genus of antibodies binding to the protein CD40CR (of unspecified species) and another claim directed to antibodies binding to the protein CD40CR expressed by activated human T cells (i.e., human CD40CR).  Noelle v. Lederman at 1346.  The application claimed priority to an earlier filed patent application which, however, did not describe any structural features of human CD40CR or of antibodies binding to it.  Id at 1346.  The priority claim was critical because of an intervening patent to Lederman disclosing an antibody to human CD40CR.  Noelle argued that since antibodies are defined by their binding affinity to target antigens, not their physical structure, he had sufficiently described human CD40CR antibody by stating that it binds to the human CD40CR antigen.  Id at 1349.  This was rejected by the Federal Circuit because of Noelle’s failure to disclose the structural elements of the antibody or antigen in the earlier application.  In rejecting Noelle’s argument, the Court expressly relied on Enzo Biochem for the proposition that as long as an applicant has disclosed a fully characterized antigen, either by its structure, formula, chemical name, or physical properties, or by depositing the protein in a public depository, the applicant could then claim an antibody by its binding affinity to that described antigen.  Id.  Indeed, the Court stated, “[i]f Noelle had sufficiently described the human form of CD40CR antigen, he could have claimed its antibody by simply stating its binding affinity for the “fully characterized” antigen.”  Id.

Centocor involved a patent (application filed in 2002), having claims to antibodies binding to human tumor necrosis factor- ? (TNF- ?), the antibodies containing human constant as well as variable regions (i.e., fully human antibodies).  Centocor at 1346.  For priority, the claims relied on earlier applications going as far back as 1994.  Id. at 1348.  The issue was whether the earlier applications provided adequate written description for the fully human anti-TNF- ? antibodies.  The earlier applications did not describe a single antibody having a human variable region.  Id.  The patentee argued that under Noelle and the PTO Guidelines, fully disclosing the human TNF-? protein provided adequate written description for any antibody that bound to it.  Id. at 1351.  The Court pointed out that PTO guidelines required not only that the antigen be fully disclosed but that generating the claimed antibody be so routine that possessing the antigen placed the applicant in possession of an antibody.  Id. at 1352.  The Court determined that as of the date of the earlier applications, making an antibody with a human variable region was not possible using conventional, routine, or well developed and mature technology, and found the claims to be invalid.  Id. at 1352, 1353.  Of note, the Court did not reject the “newly characterized antigen” test; rather it concluded that the claims failed the test, because the methods for making the claimed antibodies were not routine.

As is clear from the above, in none of these prior cases the Federal Circuit disapprove of the test.  In what appears to be a clear departure from the past, in Amgen, the Court has rejected the test, basing its rejection on the ground that it effectively eliminates the written description requirement.  Amgen at 15, 16.  According to the Court, where a functional genus claim to antibodies to a newly characterized antigen is concerned, the underlying science is not so advanced as to establish that “make and use” is equivalent to the required description of the claimed antibodies.  Id. at 16.  Drawing such equivalence, the Court said would amount to declaring a contested scientific proposition to be so settled as to be entitled to judicial notice, which the Court was not prepared to do.  Id.

The boundary between the written description and enablement requirements is not always obvious.  However, the two requirements are separate, as was underscored by the Federal Circuit in Ariad Pharmaceuticals v. Eli Lilly, 598 F.3d 1336 (Fed. Cir. 2010) (“Ariad”).  Enablement ensures that a patent informs the public how to make and use the invention of the patent without undue experimentation.  It is a consideration given by the patent owner to the public in exchange for the right to exclude the latter from practicing the invention without a license.  As such, it is a part of the quid pro quo of the patent grant.  A key role played by the written description requirement is to prevent “attempt[s] to preempt the future before it has arrived.”  Ariad at 1353, (quoting Fiers v. Revel, 984 F.2d at 1171).  The risk of preemption was a factor in Ariad, which involved genus claims covering the use of all substances capable of reducing binding of the protein NF?B to its recognition sites in the DNA of NF?B responsive genes.  Id. at 1359.  Ariad’s patent hypothesized three classes of molecules with potential to reduce NF?B activity in cells, but demonstrated none.  Id. at 1341, 1359.  Upholding the patent could have negatively impacted the future development of molecules belonging to these three classes.

However, preemption may not be a strong concern in all cases.  For example, in Amgen v. Sanofi (currently on remand), an ultimate finding of validity would expectedly hinder the future development of antibodies falling under the claimed genus.  But, would these antibodies, if made, be structurally that different from the exemplified antibodies as to be vastly superior in function?  Most likely no, given our understanding of antibody structure.  What purpose then the requirement serves in these situations?  A plausible answer seems once again to be “a consideration” given by the patent owner to the public for receiving the exclusionary right to a patent.  As stated by the Ariad majority, written description requirement is “part of the quid pro quo of the patent grant and ensures that the public receives a meaningful disclosure in exchange for being excluded from practicing an invention for a period of time.  Id. at 1354 (quoting Enzo Biochem, 323 F.3d at 970.).  As to what role might be played by “a meaningful disclosure” in the written description context, one may note Judge Newman’s comment in Ariad: “[t]he written description is the way by which the scientific/technologic information embodied in patented inventions is disseminated to the public, for addition to the body of knowledge and for use in further understanding and advance.”  Id at 1359.

If Amgen’s claims were based on structural details of the antibodies, they would most likely have survived the invalidity challenge.  This indicates that at the present time antibody technology is viewed as not being fully developed and structural information, such as an amino acid sequence, considered necessary to fully describe an antibody.  Under the patent quid pro quo, therefore, an applicant is expected to provide this information to receive a patent.


Warning & Disclaimer: The pages, articles and comments on 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 Read more.

Join the Discussion

One comment so far.

  • [Avatar for Joachim Martillo]
    Joachim Martillo
    November 22, 2017 10:16 am

    Fulfilling the requirements of 35 U.S. Code § 112 – Specification (first paragraph) can be difficult in the unpredictable arts.

    In writing the specification and providing diagrams, better to err on the side of caution with more detail to cover all possible ways the Examiner or the PTAB can reject and the CAFC can invalidate post-grant (not to mention IPR) than to take any shortcuts.

    I think we need a checklist of material that should be included in the written description for patent applications that are in the unpredictable arts like pharmaceutical technology.