Centocor v. Abbott Labs: Enforcing the Written Description Requirement in the Unpredictable Arts

In the recent Centocor Ortho Biotech, Inc. v. Abbott Laboratories (PDF), the Federal Circuit found that Defendant Abbott was not liable for patent infringement, on the basis of written description insufficiency. The Federal Circuit emphasized the patent statute’s requirement that an application must contain not only claims that define the invention, but also a written description that sufficiently discloses and conveys “possession” of the claimed invention.

About Centocor v. Abbott

The patent at issue concerns a human antibody that binds to the protein tumor necrosis factor-alpha (TNF-alpha). Overproduction of TNF-alpha has been implicated in a number of auto-immune diseases, including Alzheimer’s Disease and arthritis. The antibody, which both Centocor and Abbott claim, binds to TNF-alpha in order to regulate overproduction and reduce incidence of such auto-immune diseases. Although Centocor and Abbott have patents that claim a form of the TNF-alpha antibody, the two parties took different approaches and this resulted in antibodies with a similar “constant” region, but different “variable” regions. Centocor started by identifying a mouse form of the antibody and then modified it to create a “chimeric” protein (contains a human constant region and a mouse variable region). In contrast, Abbott pursued a fully human form, containing both a human constant region and a human variable region, from the beginning of its research.

The parties’ different research strategies contributed to the issue in this suit: whether Centocor had priority over the human form of the anti-TNF-alpha antibody. If Centocor’s patent supported the fully human form in addition to the mouse and chimeric forms, then the Court could find that Abbott’s Humira® therapy infringed Centocor’s patent. Centocor originally filed for patent in 1991, and filed a continuation-in-part (CIP) application in 1994. Although both its original and CIP applications predated Abbott’s 1996 filing, Centocor’s application focused on the mouse and chimeric antibodies. Neither of its applications contained written descriptions that disclosed a fully human antibody or an antibody with a human variable region. It was not until 2002, two years after the PTO granted Abbott a patent for the fully human form of the antibody, that Centocor added claims for a human variable region.

Findings On the Written Description Requirement

On appeal, Abbott argued that the written description in Centocor’s patent application was inadequate to convey possession of the humanized antibody. In order to comply with the written description requirement under 35 U.S.C. § 112, the applicant must convey “possession” of the invention to a person skilled in the art. Specifically here, Centocor had to prove that its written description disclosed an antibody with certain human characteristics, particularly a human variable region.

In response, Centocor asserted two arguments. First, it asserted that its claims in 2002, which referred to human variable regions, were supported by adequate written description in its 1994 CIP application. The Court rejected this argument. Upon examining the “four corners of the specification,” the Court found that the written description did not convey possession of an antibody with a human variable region. Instead, the written description only conveyed possession of a mouse variable region, and considering the state of the art at the time Centocor filed its CIP, this mouse variable region “does not serve as a stepping stone to identifying a human variable region within the scope of claims.” The specification did not describe a working example of the claimed antibody; did not disclose “any relevant identifying characteristics for such fully human antibodies or even a single human variable region”; and did not “disclose any relationship between the human TNF-alpha protein, the known mouse variable region… and potential human variable regions.” Altogether, Centocor merely asserted a “wish list” of properties of a fully human antibody without actually proving possession of one.

Second, Centocor argued that its patent fell under the “antibody exception”; in other words, that it adequately conveyed possession of the human antibody by disclosing the human TNF-alpha antigen. Centocor relied on PTO guidelines, as discussed in Noelle v. Lederman, which provide that an antibody capable of binding to a protein is adequately described “when (1) the applicant fully discloses the novel protein and (2) generating the claimed antibody is so routine that possessing the protein places the applicant in possession of an antibody.” The Federal Circuit rejected this argument. Distinguishing the present case from Noelle, the court found that (1) the protein here was not novel, and (2) obtaining the human form of the TNF-alpha antibody was not “routine” under the state of the art at the time of Centocor’s CIP application. While the Court did not hold that such a description is always insufficient, it found that here, Centocor’s possession of the human protein did not place Centocor in possession of the claimed human antibody.

In finding for Defendant Abbott, the Federal Circuit emphasized that although one can assert claims that are broader than what is disclosed in the written description, the patent only covers as much as what is disclosed. Neither the added claims in 2002 nor the disclosure of the human antigen conveyed “possession” of the claimed invention, and thus Centocor was not entitled to the human antibody.


This case speaks to the broader issue, regarding the appropriate standard for the written description requirement, particularly in the “unpredictable arts” like biotechnology and pharmaceuticals.

On one hand, patent applicants in these industries are incentivized to claim possession broadly, in terms of function. One reason applicants avoid disclosing an invention narrowly, such as in terms of structure, is that doing so creates the risk that a competitor will use knowledge of the patent owner’s invention to slightly alter its structure; because the patent is defined narrowly, the modified version would not infringe the original patent. Consequently, applicants often disclose their invention in terms of function in order to obtain broader coverage (in the present case, for example, Centocor argued that it had possession over “any antibody that [functions to] bind to human TNF-alpha.”). By making broad, “functionalized” patent claims, patent owners can protect themselves from competitors who would abuse knowledge of their patent, and they can ensure that their patent will have sufficient value for their research investment.

On the other hand, however, enforcing broad patents in fields of “unpredictable arts” creates the inevitable risk that courts will provide protection for more than what the applicant initially possessed. As Oskar Liivak argues, the “antibody exception” allows patent owners to claim more than what the applicant invented; it gives them the power to claim the “entire genus of antibodies that binds to the antigen.” Liivak argues that this is in direct conflict with the written description requirement in the patent statute, which requires that patents must be limited to claims to subject matter actually invented and disclosed.

The written description requirement poses a difficult predicament for patent applicants in the “unpredictable arts.” On one hand, a narrow disclosure greatly limits their patent protection, and on the other hand, a broad disclosure may yield their patent unenforceable in courts. There has been an ongoing struggle regarding the optimal amount of disclosure to satisfy this requirement, and while there is a persistent concern about potential abuse of narrow patents, the present case indicates the Federal Circuit’s greater concern against providing too much patent protection for overly broad patents.