There, the court found that the requirement was not satisfied for claims reciting cDNA-encoding human insulin where the application did not set forth any specific human cDNA sequences. In University of Rochester v. G.D. Searle & Co., Inc. (Fed. Cir. 2004), a case similar to Ariad, the court held invalid claims that were directed to a method of selectively inhibiting COX-2 enzymes where the application did not describe any specific compounds that exhibited selective COX-2 inhibiting activity.
In opinions dissenting from the decision to deny a petition for rehearing en banc in Rochester, Judges Newman and Dyk expressed their support of Judge Lourie’s decision applying a separate written-description requirement, while Judges Rader, Gajarsa, and Linn expressed their belief that the written-description requirement does not go beyond that required to satisfy enablement. Thus, we can predict how six of the twelve judges on the court might vote on the rehearing in Ariad, with three votes on each side of the issue.
The written-description requirement plays an important role in biotechnology patents, ensuring that the scope of patent protection is commensurate with the immediate real-world benefit the invention offers to the public. On the other hand, applying the requirement too stringently may not adequately protect pioneering inventions that have many implications and embodiments, only some of which can be nailed down by the time the patent application is filed.
Should an inventor who discovers a mechanism underlying a disease, such as the role of NF-kB, be granted exclusive rights to methods of exploiting that mechanism without disclosing any specific means for doing so? How much laboratory work should an inventor have to do before he can protect an invention with such broad implications?
There are no clear guidelines as to how much disclosure is “enough” to satisfy the written-description requirement, and no bright-line rules as to how many individual species are required to adequately describe a broad genus.
In Falko-Gunter Falkner v. Inglis (Fed. Cir. 2006), the court summarized its written-description jurisprudence as holding that “(1) examples are not necessary to support the adequacy of a written description (2) the written-description standard may be met...even where actual reduction to practice of an invention is absent; and (3) there is no per se rule that an adequate written description of an invention that involves a biological macromolecule must contain a recitation of known structure.” This enumeration of what is not always required does not provide helpful guidance on what is required.
For example, the Ariad patent mentioned three classes of molecules that could reduce NF-kB activity. Even if the patent had disclosed one example of each class of molecule, that might not have been enough to satisfy the written-description requirement for generic claims.
Three examples of one class of molecules might have been enough to support a claim directed to methods using that specific class of molecules, but even that could be subject to challenge on the basis that the examples were not representative of their genus. Would three examples of each class of molecule have been sufficient to support broad generic claims?
This uncertainty places patent applicants in the difficult position of deciding when to file their applications when the technology is still being developed. Filing early, with a few examples, carries a risk that the written-description requirement will not be satisfied. Filing later, after more work is done, carries a risk of falling behind others who may publish their work and/or file their own patent application earlier.
The same considerations do not play as large a role under the enablement requirement. This is partly because patent applicants are permitted to supplement the record to support enablement after the application has been filed, as long as the evidence is keyed to the application as filed.
For example, MPEP § 2164.05 provides that an applicant can submit a “declaration after the filing date, which demonstrates that the claimed invention works,” as long as “the steps, materials, and conditions used in the experiments of the declaration [are commensurate] with those disclosed in the application.”
Similarly, if an application is filed with data surrounding one compound, with claims directed to a broader genus, a declaration can be submitted during prosecution to show that other compounds in the genus that are described in the application, or that were identified or selected by criteria outlined in the application, exhibit the predicted activity. Thus, filing an application early is not as risky from an enablement perspective.
While the enablement requirement may represent a hurdle to claims like Ariad’s that are directed to “any” method of achieving a specific result, it often is not as significant a hurdle as the written-description requirement. Thus, abolishing the separate written-description requirement would likely lead to broader biotechnology patents. It is not clear whether that result would benefit the industry in the long run.