Base Antibody Patent Claims on Structure, Not Function, Official Says

Stay ahead of developments in federal and state health care law, regulation and transactions with timely, expert news and analysis.

April 27 — Applicants who wish to patent a group or genus of related antibodies should claim them by structure or consider claims to species differentiated by manufacturing process, a PTO speaker suggested April 26.

Joanne Hama, a member of the Patent and Trademark Office workgroup focusing on immunology and molecular biology, specifically discussed the difficulty in satisfying the written description requirement of 35 U.S.C. §112 for antibody genus claims.

“Written description is a continuum. It's easier to do the more specific you can get,” she said at the Biotechnology, Chemical and Pharmaceutical (BCP) Partnership Meeting at the PTO headquarters in Arlington, Va..

Questions after her presentation showed some persistent frustration among attendees, with one questioner asking, “Why isn't it possible to find a middle ground?”

Getting patent protection for antibodies has special importance because, according to an October 2015 report from Research and Markets, global sales for monoclonal antibodies reached $40 billion in 2014 and are expected to grow at a 10 percent rate each year through 2020. Monoclonal antibodies are made by identical immune cells that are all clones of a unique parent cell and have become an important tool in biochemistry, molecular biology and medicine.

Satisfying Written Description

Antibodies are a class of proteins that are produced by plasma cells used by the immune system to protect the body against pathogens. They attach to an antigen—a “hook” in the invading cell's surface that is partially responsible for what goes in and out of the body—to sterilize or kill the harmful substance.

Hama noted that Section 112(a) requires that the specification must contain a written description of the invention. This requirement is separate and distinct from the enablement requirement in the section, she said, citing Ariad Pharm., Inc. v. Eli Lilly & Co. (4 LSLR 261, 3/26/10).

Hama said that written description for a claimed genus may be satisfied through sufficient description of a representative number of species, and when relevant, identifying characteristics are disclosed—for example, when the prior art has established a strong correlation between structure and function. But claiming by function doesn't necessarily satisfy the written description requirement, she said, citing Regents of the University of California v. Eli Lilly & Co., 119 F.3d 1559 (Fed. Cir. 1997).

Mouse Isn't Stepping Stone

As background, Hama discussed two decisions of the U.S. Court of Appeals for the Federal Circuit: Centocor Ortho Biotech, Inc. v. Abbott Labs. (5 LSLR 224, 3/11/11) and AbbVie Deutschland GmbH v. Janssen Biotechnology, Ltd. (8 LSLR 650, 7/11/14).

In Centocor, Centocor's 1991 patent application, which was directed to a tumor necrosis factor blocker, TNF-α, disclosed and claimed a mouse antibody and a chimeric antibody, the latter making some parts of the mouse antibody more human. Abbott developed a fully human antibody and eventually received a patent, with priority to a 1996 patent application that covers its rheumatoid arthritis product Humira. After Humira was given regulatory approval in 2002, Centocor filed its claims to fully human antibodies, claiming priority to added claims it made in 1994, and received its patent in 2006. Centocor sued Abbott, alleging that Humira, which competed with Centocor's Remicade, infringed Centocor's '775 patent.

The Federal Circuit ultimately concluded there were only “limited references to fully-human antibodies, none of which relate to the specific critical limitations in the asserted claims” in the '775 patent and that, “while the patent broadly claims a class of antibodies that contain human variable regions, the specification does not describe a single antibody that satisfies the claim limitations.”

The conclusion of Centocor is that the mouse sequence didn't serve as a stepping stone for a human sequence and that a mere wish or plan for obtaining a claimed invention isn't enough, Hama said.

Enablement vs. Description

In AbbVie, AbbVie's patents had claims directed to fully human anti-IL-12 antibodies. AbbVie followed what was then the usual practice of making antibody claims on the basis of an antigen with only particular antibodies from the group disclosed and distinguished by function.

AbbVie’s patents describe more than 300 fully human antibodies that bind and neutralize IL-12. All of AbbVie's disclosed antibodies were derived from the first generation antibody Joe-9. Centocor developed Stelara (ustekinumab) for treating adults with moderate-to-severe plaque psoriasis. AbbVie sued for infringement.

A jury ruled against AbbVie, and the Federal Circuit ultimately concluded that the asserted claims “attempt to claim every fully human IL-12 antibody that would achieve a desired result, i.e., high binding affinity and neutralizing activity, and cover an antibody as different as Stelara, whereas the patents do not describe representative examples to support the full scope of the claims.”

Functionally defined claims can meet the written description requirement only if a reasonable structure-function correlation is established, the court said.

Hama stressed the difference between written description and enablement: “Written description requires description while enablement is satisfied with disclosure of how to make and use.”

Structure, Structure, Structure

In the question-and-answer period, Hama was asked, “In light of everything you've said, what would be sufficient to support genus patent claims in the antibody space?”

“Claim them by structure to make it clear what you are claiming. Or line up structures and say that these four or five or six or 10 are critical for making the claim work,” Hama said.

She added, “The first inventor is never going to get a genus. The information that's necessary comes much further down the line.” She also suggested that applicants should at least look at the option of avoiding genus claims altogether and limiting the claim to the species made by a specific process.

Another attendee asked, “How broadly applicable are Centocor and AbbVie when they are defined by the state of the art at the time?”

Hama said, “I encourage my examiners to look at the filing date of the application to see what technology was available at the date of filing.”

One questioner asked, “Is the structure/function correlation language really applicable for antibodies?”

Directors of the PTO's TC 1600 center—which focuses on biochemistry and organic chemistry and is one of the PTO's nine technology centers covering patent applications—responded that they have integrated the Centocor and AbbVie decisions into the PTO's guidelines. “We are looking at antibodies the same way we look at proteins.”

Attendees afterward told Bloomberg BNA that deciding how to patent proteins is also problematic after the Supreme Court's decision in Ass'n for Molecular Pathology v. Myriad Genetics, Inc., which invalidated isolated DNA claims and by inference similar types of claims (7 LSLR 622, 6/14/13).

To contact the reporter on this story: John T. Aquino in Washington at

To contact the editor responsible for this story: Lee Barnes at

Request Health Care on Bloomberg Law