Mayo v. Prometheus cuts back on patentable inventions

Supreme Court decision could have far-reaching effects

Many businesses and their patent counsel are upset by the Supreme Court’s unanimous decision in Mayo Collaborative Services v. Prometheus Labs because the March 20 ruling cuts back on the types of inventions that are patentable—and no one knows by how much.

“It is not really clear what the boundaries are,” says Prof. Rebecca Eisenberg of University of Michigan Law School. “There’s a lot of anxiety about what [Mayo] means.”

This anxiety is well-founded. Even a relatively narrow interpretation of Mayo could sweep away thousands of existing patents and prevent many new inventions from receiving legal protection. Medical diagnostic patents, medical treatment patents and gene patents seem particularly at risk. But Mayo’s effects may stretch much further. The decision “might apply to mechanical inventions and all sorts of other inventions,” Eisenberg says.

The case began with two patents licensed exclusively to Prometheus. These patents claimed a simple method for determining how much thiopurine to administer to a patient in order to fight certain autoimmune diseases. This method was: 1) administer thiopurine, 2) determine the resulting concentration of thiopurine metabolites in a blood sample from the patient and 3) compare this concentration with concentrations determined to be too low (and thus ineffective) and too high (producing harmful side effects). If the patient’s metabolite concentration is too low or too high, this “indicates a need to” adjust the dosage, the patents claim.

Writing for the court, Justice Stephen Breyer held that this medical treatment method was not patentable subject matter. The first two of the claimed steps—administering a drug and analyzing a blood sample—were “well understood, routine, conventional activity already engaged in by the scientific community” at the time of the patent applications.

Law of Nature

The only new thing the patents claimed was the information listed in the third step: the levels at which thiopurine metabolites were too low to be medicinally efficacious and too high to be safe. This information could constitute an important scientific discovery, but it was an unpatentable law of nature, the Supreme Court held.

This strikes some experts as an odd law of nature, because thiopurine is man-made, and there’s nothing natural about putting it into the human body. But once thiopurine is placed inside a person, nature takes over, according to the Supreme Court. The drug is metabolized by “entirely natural processes,” and those same natural processes produce the drug’s effects on a person. Thus the relationship between the amounts of thiopurine metabolites and effects on the human body is a law of nature. It is a “narrow” law with only “limited applications,” but it is a natural law, nonetheless, the court stated.

The Supreme Court has long held that laws of nature are not patentable subject matter, but Mayo greatly expanded the definition of “laws of nature,” according to many experts. “The [Mayo] court’s conception of natural phenomena and natural law is huge,” Eisenberg says. She adds, “If ‘natural law’ includes a patient’s response to a drug intervention that does not exist in nature, that could take out a lot of biotech patents.”

Many medical treatment patents and medical diagnostic patents, which are also based on a patient’s natural response to some stimulus, will also be at risk. And without such patents, many fear that the rapidly growing field of personalized medicine will be imperiled.

Such fears are overblown, according to Vincent Capuano, a partner at Duane Morris. Although Mayo puts many existing patents are at risk, the decision could have only a limited effect on patenting future inventions. “There are lots of ways to draft patents that can pass the Mayo test,” Capuano explains. The key is to claim an application of a natural law—not the law itself.

Patentable Application

Mayo reaffirmed that a natural law isn’t patentable subject matter, but an application of a natural law is. Distinguishing between the two isn’t always easy, but Mayo provided some guidelines: In order to be patent-eligible, “a process that focuses upon the use of a natural law” must contain additional “elements, sometimes referred to as an ‘inventive concept,’ sufficient to ensure that the patent … amounts to significantly more than a patent upon the natural law itself.”

The patents in Mayo failed this test. Aside from making use of a natural law, the patents claimed only two steps—administering a drug and analyzing a blood sample. Such “conventional or obvious” steps added no inventive concepts and are “normally not sufficient to transform an unpatentable law of nature into a patent-eligible application of such a law,” the court stated. These two steps, moreover, “must be taken in order to apply the [natural] laws in question,” so the patents “effectively claim the underlying laws of nature themselves,” the court wrote.

This suggests how businesses and their patent counsel can respond to Mayo. They should “review their claim language [in patents and patent applications] and ensure it is not too broad,” says James Mullen, a partner at Morrison & Foerster. “If you have a method claim, make sure you are claiming it in a relatively narrow way so you can protect your invention while not being seen as preempting the entire field.”

That’s easier said than done. Because if a claim is drafted too broadly, it may be struck down; and if a claim is drafted too narrowly, it may not protect the invention against copycats. “This creates a new tension for patent attorneys,” Mullen says.

Process vs. Product

Clever drafting, however, may not save some important patents—including patents on human genes. There are more than 3,000 such patents in the U.S., covering more than one-fifth of all human genes. And the first to be struck down may be Myriad Genetics’ controversial patents on two cancer genes.

These patents give Myriad exclusive rights to the isolated form of two human genes, BRCA1 and BRCA2, plus certain mutations of these genes. Myriad has used these patents to become the exclusive provider of diagnostic tests that determine if individuals have mutated BRCA1 or BRCA2 genes, and so are at heightened risk for breast and ovarian cancers.

In 2010, a federal district court in Manhattan struck down Myriad’s patents, holding that the genes were natural phenomena and thus not patentable subject matter. The Federal Circuit reversed, ruling in July 2011 that because the isolated forms of the genes did not exist in nature, they were patentable. The ruling was appealed to the Supreme Court, and six days after the court issued its opinion in Mayo, it vacated the Federal Circuit’s ruling and ordered the lower court to reconsider the case in light of Mayo.

The case, Association for Molecular Pathology v. Myriad Genetics, focuses on a product patent, not a process patent as in Mayo. But “the Supreme Court clearly thinks there is something to apply from Mayo,” says Prof. Arti Rai of Duke Law School.

Mayo suggests that isolated genes, such as those claimed by Myriad, are not patentable subject matter. These genes were isolated by applying conventional methods to naturally-occurring genes. This seems to lack the inventive step needed to turn an unpatentable natural phenomenon (DNA) into a patent-eligible human creation. And by claiming a monopoly on the isolated form of the gene, Myriad is asserting a monopoly on the only form of the genes that human technology can use. Its patents thus “effectively claim the underlying” natural phenomena of the genes themselves.

Nevertheless, a fair number of experts anticipate the Federal Circuit will reaffirm its prior decision and uphold Myriad’s patents. Should that happen, the case may well wind up back at the Supreme Court. Given the unanimous decision in Mayo, Myriad may not receive a friendly reception. Notes Rai, the high court’s unanimous ruling in Mayo was intended “to send a strong signal to the Federal Circuit.”

Join the Conversation

Advertisement. Closing in 15 seconds.