On July 29, 2011, a split panel of the U.S. Court of Appeals for the Federal Circuit confirmed the patentability of “isolated” DNA molecules, to the relief of many in the biotechnology community. Association for Molecular Pathology v. U.S. Patent and Trademark Office (Fed. Cir. 2011). But there was disagreement even among the two-judge majority as to the reasons why, as discussed below. Importantly, there was agreement among all three judges of the panel on the disposition of Myriad’s method claims. Myriad’s diagnostic method claims were unpatentable because they claimed only abstract mental steps, while the screening method claims were directed to patent eligible subject matter because they included process steps that were “transformative,” such as a step of growing cells and determining the growth rate of the cells. In order to write a claim that survives at least the “machine or transformation test” of patent-eligible subject matter, the majority held that at least one non-mental, “wet lab” step should be part of the claimed process. This includes, for example, a step of “administering” or “determining.” Myriad’s claims failed the test because they “do not apply the step of comparing two nucleotide sequences in a process.” Instead, “the step of comparing two DNA sequences is the entire process claimed.” Transformative process steps of, for example, extracting DNA from a sample and determining its sequence, were missing from Myriad’s diagnostic method claims.

The court rejected Myriad’s argument that these steps were implied because the term “sequence” in the claims referred to the physical DNA molecule (which would require that its sequence be determined before the sequences could be compared). The court found instead that the term “sequence” referred not only to a DNA molecule but also to its information content, i.e., the linear sequence of nucleotides AGCT. Accordingly, the comparison of two sequences could be accomplished by “mere inspection” and the claims failed the machine or transformation test. In contrast, Myriad’s claim to a screening method included the mental step of “comparing” the growth rate of cells in a process that also required the transformative wet lab steps of “growing” the cells and “determining” their growth rates. The inclusion of these transformative steps made the screening methods patentable.

Regarding the patent-eligibility of isolated DNA molecules, the court’s inquiry focused on whether isolated DNA fell within the “product of nature” exception to patentable subject matter established by the U.S. Supreme Court in Diamond v. Chakrabarty, 447 U.S. 303 (1980) and Funk Brothers Seed Co. v. Kalo Inoculant Co., 333 U.S. 127 (1948). There was general agreement on the panel that the test is whether the “human-made invention” of isolated DNA is “markedly different” from the native DNA that exists in nature. But all three judges differed in their opinion of what is required for the human-made composition to be considered “markedly different” from what exists in nature.

In Judge Lourie’s view, the chemical differences between an isolated DNA molecule and the DNA polymer as it exists in nature in the form of a chromosome are sufficient to make the isolated molecule markedly different. Writing for the majority, Judge Lourie held that isolated DNA is markedly different because the isolated DNA molecules “have a distinctive chemical identity and nature from molecules that exist in nature.” Judge Lourie found unpersuasive plaintiffs’ argument that the similarity (indeed, even the exact identity) of the linear sequence of nucleotides in both the isolated and native DNA molecules compelled a finding that they are not markedly different. Instead, in Judge Lourie’s view, “[t]he claimed isolated DNA molecules are distinct from their natural existence as portions of larger entities, and their informational content is irrelevant to that fact.”

In her concurring opinion, Judge Moore also found claims to an isolated DNA molecule patentable, but for a different reason. To her, the literal chemical differences between an isolated DNA molecule that includes most or all of a gene and the gene itself as it exists in nature, i.e., as part of a chromosome, were insufficient to confer patentability: “Although the different chemical structure does suggest that the claimed DNA is not a product of nature, I do not think this difference alone necessarily makes isolated DNA so ‘markedly different’ from chromosomal DNA as to be per se patentable.” Judge Moore’s analysis turned instead on whether the differences in structure “impart a new utility which makes the molecules markedly different from nature.” Judge Moore had little difficulty finding smaller molecules patentable because they have “markedly different properties which are directly responsible for their new and significant utility,” for example, as probes or primers. And she had even less difficulty finding cDNA molecules patentable because they do not exist in nature at all and therefore cannot fall within the “laws of nature” exception to patentable subject matter. But in Judge Moore’s analysis, a different outcome was warranted for longer molecules encompassing most or all of the native gene: “Despite the literal chemical difference, the isolated full length gene does not clearly have a new utility and appears to simply serve the same ends devised by nature, namely to act as a gene encoding a protein sequence.” Instead, Judge Moore concurs in the judgment on this point out of deference to Congress and her view that the “settled expectations” of the patent community require it.

Judge Bryson, in dissent, used an analysis similar to the structure/function framework of Judge Moore, but came to a different conclusion. Judge Bryson would have upheld only the cDNA claims, finding no rationale in the law or precedent for upholding claims to genes or “gene segments.” There is thus significant disagreement among this three-judge panel about the patentability of isolated DNA molecules that encompass most or all of a gene. And perhaps surprisingly, two of the three judges agree that there is no compelling rationale in the law for finding these molecules patentable. But for Judge Moore’s deference to “settled expectations” and the role of Congress, DNA molecules encompassing entire genes could have been excluded from patentability.

In summary, pending the outcome of any appeal of this decision, it would be prudent to ensure that claims to smaller subsequences having a definite utility as well as claims to cDNA molecules are included in any patent application seeking to cover similar technology. And patent practitioners would do well to ensure that their diagnostic and screening method claims include at least one clearly non-mental step of “determining” something that is of central importance to the claimed method. While adding such a step may make it more difficult to enforce the claim, this seems to be the tradeoff required to satisfy the machine-or-transformation test of patent eligibility. Although this test is not the only test for patent eligibility of method claims, it is, for now, the only one consistently articulated and applied by the Federal Circuit.