Myriad, Finally: Supreme Court Surprises by not Surprising

After what seemed like an eternity, the epic saga known as AMP v. Myriad Genetics has finally come to a close. On June 13, 2013, the Supreme Court ruled (1) that isolated genomic DNA (gDNA) is not patent-eligible under section 101 of the Patent Act, but (2) cDNA is. For once, what the Justices said at oral argument gave accurate clues to what they really thought, and the result was what almost every observer (including this one) had predicted.

The opinion—by Justice Thomas—was unanimous and brief: 18 pages. Justice Scalia wrote a one-paragraph concurrence in which he said only that he didn’t know enough to sign on to the Court’s recitation of “the details of molecular biology,” though he agreed with the decision. The 9-0 decision reflects a recent trend in patent decisions (see Mayo v. Prometheus and last month’s Monsanto opinion), and may represent a reaction to the derision that greeted the fractured, multi-opinion mess that was Bilski v. Kappos, the Court’s 2010 business methods case. Beyond the surface unanimity, the Myriad opinion has a no-brainer quality—that is, the opinion does not give any sense of wrestling with a difficult issue, giving the impression that the outcome was so clear that reasonable minds couldn’t possibly differ. Consistent with this approach, the Court cited only the lower court opinions and briefs in the case, a few of its own patent decisions, and a single scientific text (Watson’s Molecular Biology of the Gene). No amicus briefs (except the government’s, to confirm that the Department of Justice had thrown the USPTO under the bus), no scientific or law review articles.

The strict legal significance of the decision is pretty straightforward. The Court struck down patent claims on genomic DNA that has been merely “isolated” from the body, where “isolated” means, well, “isolated”—removed and separated from its natural environment in the cell. More specifically, the Court held that genomic DNA does not meet the threshold test of patentable subject matter under section 101. It upheld the subject matter status of cDNA, which it defined as “synthetically created DNA . . . which contains the same protein-coding information found in a segment of natural DNA but omits portions within the DNA segment that do not code for proteins [introns].” Every patent drafted like Myriad’s first and broadest claim—“an isolated DNA coding for [a specified protein]”—is now invalid. Conversely, claims limited to cDNA versions of genes continue to pass the threshold test, though they are still subject to scrutiny under all the other patentability requirements (more on that below).

The cDNA/gDNA distinction has its roots in the Federal Circuit’s 1991 decision in Amgen v. Chugai. That case turned on a priority-of-invention contest: who got the rights to the human EPO gene. But the case did, at least implicitly, establish the subject matter status of cDNA. The broadest claim upheld in that case was: “A purified and isolated DNA sequence consisting essentially of a DNA sequence encoding human erythropoietin.” The court treated this as a claim to the cDNA version of the gene, interpreting “purified” as meaning something like “only the coding regions.” It commented that “[i]t is important to recognize that [neither competing inventor] invented EPO or the EPO gene. The subject matter [of this claim] was the novel purified and isolated sequence which codes for EPO.”

Thereafter, two things happened. First, the subject matter eligibility of cDNA was never again challenged, until now. And understandably so, if DNA is viewed as a chemical for patent law purposes. cDNA is, in a literal sense, a substance not found in the body; it contains the same coding information as the mature RNA transcript, but it is a different chemical: and, with the noncoding regions excised, it is not exactly replicated in natural DNA. The second thing was less predictable. Amgen set up one of those slippery slopes that lawyers are always talking about. The USPTO ignored Amgen’s “important to recognize” comment, as well as its implied cDNA/gDNA distinction, and allowed claims like Myriad’s, with the only limitation being “isolated.” Until now, no case focused on this distinction. One way to look at what the Supreme Court held is that it has simply reaffirmed what the Federal Circuit’s said in Amgen, that: Yes, those kinds of claims—isolated and purified—are still OK, but extending patentable subject matter to merely isolated gDNA wasn’t contemplated by Amgen, and goes too far.

Who is really affected, and how? One way to assess the practical significance of the decision is to look at it from the perspective of various affected parties. The first and most obvious is Myriad Genetics itself. (And we can expand this category to include all companies whose business models are based on patents on single genes in genomic DNA form—although I don’t think there are many.) The Myriad stock price has taken a significant hit. But Myriad itself has been taking a consistent line on the relative unimportance of its genomic DNA patents, and it’s probably right. First, as it quickly pointed out on the 13th, about three-quarters of its BRCA-related patents haven’t been invalidated, including claims to cDNA, probes, and some of its methods. And in any event, the invalidated patents would have begun to expire in the next year or so. But more significantly, as it stressed in announcing the expansion of its European business, its most valuable asset may be the proprietary database it has built up over the life of its patent-based testing monopoly. This database consists of test results—DNA sequences—and associated health outcomes, and gives Myriad a unique advantage in interpreting mutations, particularly the lesser-known ones (variants of unknown significance, or VUS). This advantage is entirely unaffected by the decision, and will remain in place until Myriad decides to make its database public or others manage to replicate it, which won’t happen any time soon.

That leads us to the next constituency: potential direct competitors of Myriad in the breast and ovarian cancer testing market. To the extent that other companies wanted to get into that market by using the patented genomic DNA in isolation, they were barred by Myriad’s patents. Now they’re not. But the patents that barred them were due to start expiring soon, anyway. Presumably, prospective competitors were gearing up for those expiration dates. With this decision, they can move a little faster. But are they prepared to do so? How soon will it take competitors to get their technology, regulatory approvals, marketing strategies, and the like in place? And don’t forget Myriad’s advantage in interpreting results. Will doctors and insurance companies trust new market entrants, at least in the short run? I wouldn’t expect prices for BRCA testing to drop quickly.

What about the patients on whom Myriad has imposed its testing monopoly? For the reasons just stated, I wouldn’t expect the financial burden they face to change in the short term. One important thing that should happen, though, is more ready availability of second-opinion testing. For the most part, patients with positive results who wanted a second test had to go back to Myriad—hardly a “second opinion.” Competitors may well enter this market quickly. In addition, many hospitals were apparently doing second-opinion testing in-house, relying on Myriad ignoring them (which it did, with some exceptions). Now they can come out into the open.

Basic researchers are now also free to do whatever they want with isolated genomic DNA. Remember that in this country there is no meaningful research exception to patent infringement, even in the nonprofit sector; unauthorized use is infringement, period. There have been several studies of whether gene patents cast a shadow on basic research, but I find it inconclusive and unpersuasive. I have never talked to a researcher who felt inhibited—to the point of not doing a project—by the prospect of patent infringement. (And see the related ACMG discussion below)

Next, consider the impact on medicine more generally. The key segment here is genetically-based personalized medicine, which uses gene sequences to tailor drugs and treatments for cancer and other diseases to the genotypes of individual patients. Companies in this field have been well aware that many of the genes they would be sequencing would likely to be the subject of patent claims. They have also largely concluded that identifying relevant patent owners and negotiating licenses would be logistically and financially impossible. First, you’d have to find them. Then, if you did: patent license royalties are typically about 5% of revenues. If you wanted to sequence 20 patented genes, that would be 100% of your revenue. Not a promising business model.

The personalized medicine industry is still more concept than reality. But I’ve seen no evidence of current or prospective players being deterred by this “patent thicket.” Instead, I’ve heard about companies forging ahead in reliance on two beliefs. The first is that patent owners wouldn’t sue them, because of some combination of (1) lack of will or wherewithal (remember that the overwhelming majority of patents just sit there, unpracticed and unexploited); (2) fear of exposing their patents to litigation and the attendant possibility of invalidation (ask Myriad about that one); and (3) at least in the case of large companies, reluctance to take the public relations hit of being seen as standing in the way of “curing cancer.”

The second belief is technical: that so-called next-gen sequencing methods (a misnomer, since they’ve now been around for several years) have bypassed Myriad-style claims because they don’t actually use entire genes in isolation. Remember that to infringe, you must use the invention, exactly as described in a patent claim. In the case of a claim to a single isolated gene, in genomic DNA form, you must use precisely that. I’m not a geneticist, so I can’t pass judgment on that argument, but it is one that I hear regularly from scientists following this case.

In this context, it’s interesting to look at the recently, and already controversial, American College of Medical Genetics Recommendations for Reporting of Incidental Findings in Clinical Exome and Genome Sequencing. In a nutshell, the ACMG recommends that doctors and labs doing any genetic testing should always sequence 57 listed genes and report the results to the patient, regardless of whether the patient wants those results. No wonder it’s controversial. But what is interesting from the Myriad perspective is that this is all that the ACMG report says about patents: “We also did not address issues of patents in making these recommendations.” In other words, the leading organization of medical geneticists has issued a recommendation—which could quickly become the standard of care—that its members always sequence 57 listed genes without worrying about whether those genes are patented. The ACMG appears simply to ignore the gene patent thicket.

Some final legal thoughts: First, as I’ve pointed out in the past, the patentable subject matter dispute is just that, a dispute over whether genes clear the very first hurdle on the road to patentability. But even if a gene claim gets over that hurdle, it still has to clear other, potentially formidable barriers, especially nonobviousness. In 2009, in an underpublicized decision called In re Kubin, the Court of Appeals for the Federal Circuit (which hears all patent appeals and is the last stop for almost all patent cases) significantly tightened the nonobviousness requirement for gene patents. Reversing previous law, it held that (1) knowing a lot about the protein a target gene encodes plus (2) general knowledge of the techniques for isolating and sequencing target genes renders the target gene itself obvious. This holding has, since 2009, made it considerably harder to get single-gene patents in any form. Thus, while Myriad may throw out a lot of older gene patents that date from a time when less was obvious, it is questionable whether there would have been many new Myriad-style patents that would have survived the Kubin standard. That is, many or most of the single gene patents that Myriad has foreclosed would have gone down on obviousness grounds anyway. Kubin is likely having a similar effect on cDNA patents as well, undercutting the significance of the Court allowing them to survive as patentable subject matter.

My last thought concerns the scope of the decision. The Court was careful to point out what it did not decide, including the status of biotech method claims (already significantly limited by the Court’s Mayo v. Prometheus decision and the Federal Circuit’s methods holdings in Myriad, which were not before the Supreme Court), applications of knowledge about genes, and “the patentability of DNA in which the order of naturally occurring nucleotides has been altered.” I would add that the court left unaddressed the gray area of claims that fall into neither the “isolated” gDNA nor cDNA, such as gDNA that has been “purified” or altered in some other way. But notwithstanding these disclaimers, I wonder about how the lower courts and the USPTO will apply the decision to other isolated bodily substances such as proteins (where there are patents based solely on isolation) and cell lines, which begin with the isolation of the cell from the body. This may become a major issue in the future, and we will track it as it develops.

Did the Court get it right? In legal writing, the phrase “the better view” is code for “my view.” All things considered, I think the Court adopted the better view in Myriad.