The ACLU rarely involves itself with patent cases, so it was surprising when the civil liberties group filed an amicus brief urging the Supreme Court to invalidate a patent in the case of Mayo v. Prometheus. Mainstream press coverage portrayed the dispute between two medical testing labs as a mundane argument over “individualized medicine,” but to the ACLU, what’s at stake is nothing less than the freedom of thought. “What Prometheus seeks to monopolize through its patents,” the group wrote, is the “right to think” about a particular correlation between blood chemistry and the optimal dosage of a drug for treating autoimmune disorders.
Whether or not you consider Prometheus’ patent as a form of thought-policing, a Supreme Court decision to uphold it—and thereby legalize medical diagnostic patents more generally—would be a disaster for patients. It would allow third parties to be sued for simply providing doctors with accurate scientific information, and, even worse, it would set a precedent opening up the nation’s accountants, lawyers, computer programmers, and other white-collar workers to their own frivolous patent lawsuits.
Patent law has traditionally offered protection for a wide variety of machines and physical processes, but not for purely intellectual inventions. “Abstract ideas are not patentable subject matter,” Justice Stevens wrote in 1978. “A patent could not issue, in other words, on the law of gravity, or the multiplication tables, or the phenomena of magnetism, or the fact that water at sea level boils at 100 degrees centigrade and freezes at zero—even though newly discovered.”
For decades, patents on computer software straddled the boundary between the patentable physical world (a new kind of microprocessor, let’s say) and the unpatentable world of the mind. A computer program is a collection of algorithms, and two Supreme Court decisions in the 1970s held that these sorts of abstractions—such as a method for converting between binary number formats—were not eligible for patent protection. Yet when a bunch of algorithms are loaded onto a computer, they can have effects on the real world. In 1981, the Supreme Court applied this reasoning to allow a patent on a software-controlled rubber-curing machine, ruling that the mere presence of software did not rule out patent protection for an invention that otherwise fit comfortably into a traditional patent class.
Attorneys seized on this latter decision and used it to erode the traditional rule against software patents. In the 1990s, what began as a gradual process became more abrupt. A company called Signature Financial Group had obtained a patent on software for managing a mutual fund, prompting a legal challenge from a competitor, State Street Bank. In 1998, the case reached the U.S. Court of Appeals for the Federal Circuit, which has jurisdiction over patent appeals. The Federal Circuit not only upheld Signature’s patent, but it did so in sweeping terms, ruling that “the transformation of data, representing discrete dollar amounts … constitutes a practical application of a mathematical algorithm.” This reasoning arguably contradicted the relevant precedents, but the Supreme Court seems not to have anticipated the decision’s dramatic consequences and chose not to review it.
That case didn’t just remove the remaining limits on software patents, it opened the door to patents on investment strategies, tax preparation methods, and other purely intellectual activities. After all, you don’t need a computer to “transform data.” People can do that (and infringe such patents) with a pencil and paper, or even by doing the wrong calculations in their heads.
Health care was one of the first industries to be affected. Patents have long been available for drugs and medical devices, but the State Street case unleashed a new generation. In 2002, the U.S. Patent and Trademark Office issued a patent on a way to treat autoimmune disorders such as Crohn’s disease. When a patient takes a class of drugs called thiopurines, the human body converts the active ingredient into a related chemical called a metabolite, and the concentration of these metabolites can be measured in the patient’s blood. The two researchers in Montreal who applied for the patent had not invented a new drug, nor had they devised a new method for measuring metabolite levels. They hadn’t even invented the idea of calibrating drug dosage based on metabolite levels. Rather, their patent claimed that they’d come up with specific ranges of metabolite levels that would “indicate a need” to raise or lower drug dosages.
The patent wound up in the hands of Prometheus Laboratories, a firm that sells a product for testing the level of thiopurine metabolites. When a branch of the Mayo Clinic announced its own competing thiopurine testing product, Prometheus sued. Mayo’s test product, the company argued, would also give recommendations about whether the measured levels “indicated a need” to raise or lower the drug dosage. Providing this information to the doctor would cause the doctor to think about the scientific correlations described in the Prometheus patent, and that such thoughts constitute infringement of the patent.
It’s hard to think of a clearer example of the kind of “abstract idea” the Supreme Court ruled out of bounds decades ago, or of a way in which such lawsuits would improve medical care instead of harming it.
Rather than suing doctors directly, Prometheus’ strategy has been to sue third parties—like Mayo—who help doctors infringe its patents. As a brief from the American Medical Association points out, such lawsuits still harm the quality of medical care, because they encourage third parties like the Mayo Clinic to withhold relevant scientific information from doctors.
But in oral arguments on Dec. 7, no one brought up the AMA’s concerns about access to scientific knowledge nor the ACLU’s concerns about freedom of thought. Indeed, Justice Kagan wondered aloud why Prometheus didn’t file a patent that said, “at a certain number you should use a certain treatment, at another number you should use another treatment.” Such a patent, she declared, “clearly would have been patentable. Everybody agrees with that.”
(In fact, a number of public interest organizations have told the court they do not agree with that.)
The possibility of the Supreme Court allowing patents on medical decision-making should alarm even those who have little contact with the health care industry. Numerous industries that have traditionally been spared the costs of patent litigation could suddenly be forced to lawyer up. An accountant could patent the idea that a particular pattern of financial transactions “indicates a need” for a more thorough audit. A chef could patent the idea that a particular temperature “indicates a need” to take a dish out of the oven.
These ideas may sound absurd, but patents on medical decision-making would have sounded equally absurd 15 years ago. Yet the Federal Circuit’s reasoning in Mayo suggest that people could get such patents with a bit of clever lawyering, just as they did with the Prometheus patent.
By allowing the 1998 ruling on financial software to stand for more than a decade, the Supreme Court has created a dangerous confusion over what sorts of ideas are eligible for patent protection. Attorneys have taken full advantage, pushing patent law into new industries. Investment strategies were first; now medicine; eventually maybe accounting or even baking. The longer the Supreme Court lets the uncertainty linger, the more disruptive it will be to change course.
This might not be so disturbing if the patent system were serving these newly patent-eligible industries well. Unfortunately, the most thorough empirical research suggests that the cost of litigation in software and finance dwarfs the rewards the patent system provides to innovators in those industries. The patent system seems to be actually discouraging innovation in these industries.
Yet the justices of the Supreme Court did not seem to grasp the urgency of the case. The oral arguments focused more on which kinds of medical diagnostic patents should be eligible for protection than on whether such patents should be allowed in the first place. So even if the Prometheus patent is rejected, it’s likely to be a narrow ruling that allows the trend toward patents in more and more industries to continue.