When the Grass Eats the Cows

Back in 2003, one of us (John) and patent lawyer Roberte Makowski published an article in the Journal of the Patent & Trademark Office Society in which we argued that litigants and courts should revisit whether patents on genes, as conventionally drafted, violated the prohibition on patenting products of nature. We noted that claims on "isolated" (removed from their natural environment in the body) or "purified" (often, with non-coding regions removed, as with cDNA) genes evaded this prohibition by putting the focus entirely on chemical differences between such genes and their naturally occurring precursors. We then asked whether this exclusive focus made sense, given that people are interested in patenting genes because of their information-carrying capacity, which is the same in the natural and patented versions. We also noted that neither the United States Patent Office nor the courts had ever squarely addressed the product of nature argument. The USPTO had brushed it aside with references to novelty and utility, which are separate statutory questions, and the single court decision to consider it even obliquely-the Federal Circuit's 1991 decision in Amgen v. Chugai-had dismissed it in a footnote. Chemistry triumphed over information, and the semantics of patent claiming seemed to have foreclosed a potentially fundamental issue.

Then came Judge Sweet's Myriad (Association for Molecular Pathology v. USPTO) district court opinion invalidating Myriad's isolated gene patents and, at least for a moment, the world turned upside down (the title of the song that Cornwallis' band played during the surrender at Yorktown: "If ponies rode men and if grass ate the cows"). Judge Sweet adopted the information-first argument, deriding the gene-in-isolation distinction as "a patent lawyer's trick." Roberte and John were simultaneously gratified and horrified that their thought experiment had become (at least ephemeral) legal reality.

As we await the Federal Circuit's decision in Myriad, there's very little new to say about the case. Reactions have been predictable. Disinterested legal commentators have almost universally predicted that the Federal Circuit will reverse or at least significantly limit the scope of the district court's ruling-it's just too doctrinally radical to stand, posing too much risk to the biotechnology industry. Advocates have performed their expected roles: Big Biotech sees signs of the apocalypse, while public domain enthusiasts see a new golden age of genetic research unencumbered by patents.

Most of this attention, however, has been focused on what might be the less important part of the case: the gene patents. Technology and other principles of patent law may be conspiring to reduce the practical significance of single-gene patents. On the technology side, more efficient (and much cheaper) whole-genome sequencing may permit new applications to avoid "using" protected single genes in isolation. (The Patent Act prohibits making, using, or selling the patented invention exactly as it's claimed in the patent.) Nonetheless, at least in the short to medium term, single-gene patents are likely to impede the development of gene-based diagnostic techniques (especially "multiplex" techniques that use multiple genes).

On the legal front, note that Myriad is about only one aspect of patentability: whether the claimed invention is patentable subject matter. To say, as Judge Sweet did, that an isolated gene is a product of nature is to say that it does not fall within the realm of patentable subject matter. But in a 2009 case called In re Kubin, the Federal Circuit made it much easier to defeat single-gene patents on obviousness ground. Reversing its own 1995 precedent, the court held that a gene (in cDNA form) was obvious when a lot was known about the protein it encodes and the method for sequencing it was also well-known-in other words, when it was "obvious to try" to sequence that gene. This reinterpreted obviousness requirement may prove to be an insurmountable barrier to many, if not most, single-gene patents in the future.

The less-discussed but potentially more significant patents at issue in Myriad are the company's method patents. These cover methods of making sense of gene test results and using them to make diagnostic predictions. They are extraordinarily broad, allowing Myriad to monopolize just about any diagnostic use of the breast cancer genes, regardless of the patents on the genes themselves.

The patentable subject matter status of methods generally is up in the air. The Supreme Court had a chance to clarify things this past summer in Bilski v. Kappos, which involved a simple business method, but its fractured opinions only confused things further. All we know for sure is that the Federal Circuit's machine-or-transformation test (which requires that a method be tied to a particular machine, or transform something into a different state) cannot be the sole test. We await the Federal Circuit's application of Bilski, as well in Prometheus v. Mayo, which involves another simple method, this time for adjusting the dose of medication after measuring its level in the body. The Federal Circuit's decision in Prometheus and another similar case (Classen v. Biogen IDEC) will presumably shed some light on the likely fate of Myriad's method claims, and on medical association and diagnostic patents more generally.

Amidst all this legal uncertainty, it is worth stepping back and taking a practical look at the whole situation. Both sets of extreme reactions to Judge Sweet's Myriad decision are almost certainly wrong: it is not the end of the world for commercial biotechnology, nor is it the best possible outcome for science and medicine and their constituents-that is, us. Gene patents themselves are not evil. On the contrary, they-and their promise of monopoly rents-have played an essential role in financing the biotechnology industry. The problem is not with gene patents per se, but with the way that they can be used as barriers to research and innovation. If Myriad had made different business judgments, its patents would probably not be under attack. Would the ACLU have targeted Myriad if, for example, it permitted second-opinion testing by independent labs at patient-friendly prices? Or if it had announced a broad, royalty-free research license for its genes?

The complexity of the gene patent landscape-keep in mind that there are patents on an estimated 20% of human genes-along with the actions of certain patent holders, including Myriad, have combined to produce a dense gene patent "thicket" that threatens biotechnology innovation, particularly in the area of multiplex diagnostic technologies. The public disaffection with gene patents that Judge Sweet's opinion expresses so clearly will only get worse if the gene patent thicket begins to materially impair the development of these technologies. It's time for the biotechnology industry to move toward the kind of patent pooling that has facilitated the advance of so many other technologies, from sewing machines to jet engines. Innovators should have a place for one-stop shopping at reasonable prices, rather than facing individual patent-holders bent on monopolization and stacked gene patent royalties that threaten to exceed 100% of their revenues. Some biotech players might have short-term reasons for opposing this approach, but the current legal mess ought to be a wake-up call: if the industry doesn't move in this direction of a rational economic solution, courts and legislators will likely impose something far worse.