The recent case of UCB, Inc v Yeda Research and Development Co, Ltd demonstrates the importance that prosecution history estoppel can have on claim construction. While the estoppel effect is generally understood with regard to claims that were actually amended during prosecution, the doctrine can also be used to reduce the breadth of other claims, even those that were not amended.

Yeda Research and Development Company owned a patent directed to a monoclonal antibody that selectively binds and purifies the human cytotoxin, tumor necrosis factor (TNF). UCB, Inc produces a pharmaceutical agent, Cimzia®, which is a humanised monoclonal antibody that treats various immunological disorders by binding active TNF to neutralise TNF's cytotoxic effects. The lower court declared that UCB's Cimzia product did not infringe on Yeda's existing patent.

In affirming the declaration, the Federal Circuit relied heavily on the doctrine of prosecution history estoppel, which forbids a patent owner from recapturing subject matter that was cancelled or amended during prosecution if the cancelation or amendment was made in response to a rejection by the examiner. Although prosecution history estoppel typically affects construction of claims that were actually amended during prosecution of a patent application, the Federal Circuit used the doctrine to drastically reduce the breadth of Yeda's facially strong patent, even when the claim at issue was not so amended.

During prosecution of the patent application, Yeda limited all claims to a mouse (murine) monoclonal antibody to distinguish Yeda's invention from prior art covering a rabbit monoclonal antibody. However, more than 10 years into prosecution, Yeda added new claims to regain the broader, originally filed term 'monoclonal antibody', arguing that the recitation of "murine" in the claims was unduly limiting. Since a mouse monoclonal antibody was the only species expressly cited in the specification, the examiner initially rejected these new claims for lack of enablement. Yeda's response relied heavily on a prior publication and an accompanying expert declaration asserting that, at the time of the patent application, a person having ordinary skill in the art could produce monoclonal antibodies from species other than mice. The declaration also stated that, at the time of the application, a person having ordinary skill in the art could genetically engineer chimeric antibodies, which are antibodies comprising genetic material from more than one species. In response to Yeda's argument, the examiner removed the rejection and allowed a claim broadly directed to a monoclonal antibody.

However, Yeda went a step further and submitted two more new claims that depended on the "monoclonal antibody" claim. These new dependent claims expressly recited antibodies made from other species (eg, rat, hamster and human) and recited mouse chimeric monoclonal antibodies, which includes humanised antibodies. The examiner rejected these new claims for introduction of new matter that was not supported by the specification. In response to the rejection, Yeda cancelled the two new claims, but kept the broader, independent claim. A patent soon issued with claim 1 broadly directed to a "monoclonal antibody".

At trial against UCB, Yeda argued that 'monoclonal antibody' is a generic term and, as such, should be construed according to its broadest reasonable interpretation in light of a person having ordinary skill in the art's knowledge at the time of the application. However, the court applied the general rule of prosecution history estoppel to considerably restrict the claim's interpretation. The court interpreted Yeda's cancellation of the two dependent claims during prosecution as a concession that the broadly worded independent claim to a monoclonal antibody did not include chimeric and humanised antibodies. Therefore, while Yeda's cancelled claims were narrower than the claim that was actually allowed, canceling the narrower claims paradoxically restricted the scope of the broader claim – even though the allowed claim had not been so amended. Consequently, Yeda was left with a facially broad patent that, when interpreted in light of the prosecution history, was narrowly directed to a mouse monoclonal antibody. Thus, UCB's humanised monoclonal antibody did not infringe Yeda's existing patent.

Comment

Yeda's misfortune provides a valuable lesson for both patent applicants and those attempting to build a strong patent portfolio: do not underestimate the effect that prosecution history may have on the value of a patent. Before purchasing a patent, the prospective buyer needs to 'look under the hood'. Dig deep into prosecution history and read office action responses with a keenly critical eye to ensure that the patent is not a lemon. Importantly, the buyer should not assume that the prosecution history will not negatively affect claims that were not amended. If the applicant narrowed claims during prosecution in response to an examiner rejection, make a conscious effort to confirm that the narrowing did not restrict construction of related claims that are particularly important to the purchasing business or are integral to the scope of the invention.

Similarly, when prosecuting a patent, the applicant should be mindful of how any concessions, even in dependent claims, could weaken the patent as a whole. As a safeguard, it may be advisable to resist going for the entire carrot in a single bite. If possible, take the broader claims that the examiner allows and utilise the continuation process to further refine the claims. In hindsight, Yeda may have prevented the 'broadest reasonable interpretation' argument from being completely swallowed by prosecution history estoppel by simply accepting the broader claim that was allowed and moving the battle for the species claims to a continuation application. At the very least, the continuation process would have provided Yeda additional time to craft a well-supported argument for allowance of the species claims.

This case highlights the potentially broad, sweeping impact that prosecution history may have on the interpretation of not only amended claims in a patent, but the patent as a whole. Through the pervasive effects of prosecution history estoppel, a facially well-crafted patent may, in fact, provide only razor thin protection. Although it may take more time and cost a bit more money upfront, a detailed review and consideration of the prosecution history could provide invaluable insulation to prevent a virtually worthless investment.

Aaron Chaloner

This article first appeared in IAM. For further information please visit www.iam-media.com.