Friday 23 March 2012

Mayo v. Prometheus: US Supreme Court struggles to distinguish “laws of nature” from patentable inventions


The U.S. Supreme Court’s new decision in Mayo Collaborative Services (dba Mayo Medical Laboratories) v. Prometheus Laboratories, Inc., Case No. 10-1150 (U.S. Mar. 20, 2012), adds another piece to the puzzle of evaluating patent eligibility under 35 U.S.C. § 101. The 2010 decision in Bilski v. Kappos, ___ U.S. ___, 130 S.Ct. 3218 (2010) laid out a framework for identifying unpatentable “abstract ideas.” In Mayo, the Court examines “laws of nature.”

In a 9-0 ruling, the Court held that Prometheus Laboratories, Inc.’s patents covering methods for adjusting doses of thiopurine drugs in patients with autoimmune disease were not patent-eligible. The Court ruled that the physiological relationship cited in the claims – that thiopurine metabolite concentrations below 230 pmol per 8x108 red blood cells are ineffective, but concentrations in excess of 400 pmol per 8x108 red blood cells are toxic – is a “law of nature.” As a result, the court ruled, a claim directed to a method utilizing that relationship must be restricted to an application of the relationship, not the law of nature itself.

Although the Prometheus patents recited steps of “administering” thiopurine drugs and “determining” the metabolite concentrations in a patient’s blood, thereby indicating whether the patient dosage was too high or too law, the Court ruled that the claims merely covered the law of nature itself. The Court discussed two factors in determining eligibility: (1) whether the claim recited features, such as unconventional or non-routine steps or structures, that provide “practical assurance that the process is more than a drafting effort designed to monopolize the law of nature itself[;]” and (2) whether the claim would “risk disproportionately tying up the use of the underlying natural laws[.]” Mayo, at 4, 8-9. The Court ruled that the steps of administering thiopurine to patents and then determining metabolite blood concentrations were well-known, routine, and obvious. Thus, they did not restrict the claim to an application of the underlying laws of nature. In addition, the Court ruled that the claim would disproportionally preempt innovation and research in the field, because the only way to exploit the law of nature (at least in the Court’s eyes) was to practice the steps recited in the claim.

Although the effect of this decision will depend on how it is applied by the lower courts, including the U.S. Court of Appeals for the Federal Circuit, the decision raises some grave issues concerning the eligibility of process claims, especially in medical diagnostic process inventions.

First, Court’s assumption that the precise relationship between metabolite concentration and effectiveness/toxicity is a law of nature seems suspect. Every scientific or technical invention will necessarily depend of some physical law. One would think that laws of nature were broader scientific relationships, such as the law of gravity, the mechanisms of DNA replication, or laws of physics and thermodynamics. The specific relationship recited in the Prometheus claims – including specified thresholds for drug effectiveness and toxicity – would appear to be an application some laws of physiology, not a law of nature in and of itself. If the fact that thiopurine metabolite concentrations in excess of 400 pmol per 8x108 red blood cells is toxic is a law of a nature, what scientific fact would not be a law of nature? In effect, most – if not all- process claims could be subjected to this standard.

Second, the apparent requirement that a claim must recite unconventional, non-routine and unobvious steps or structures in order to be restricted to an “application” of a natural law seems unsupportable. The Court justified its rule in part on its milestone decision in Diamond v. Diehr, 450 U.S. 175 (1981). The process invention in Diehr employed the well-known Arrhenius equation, governing the curing or rubber, and recited steps of installing uncured rubber into a press, closing the mold, constantly determining the temperature, constantly recalculating the cure time using the equation, and opening the press. The Court concluded that those steps, unlike the steps recited in the Prometheus patents, were not conventional or routine. But it would appear that the steps would have been common in the rubber curing art.

Third, the decision calls into question the Court’s statement in Bilski that the “Machine or Transformation” test is “a useful and important clue, an investigative tool, for determining whether some claimed inventions are [eligible] processes under § 101[.]” Bilski, 130 S.Ct. at 3227. The Federal Circuit had held that since the thiopurine metabolites resulted from chemical transformations in the patient’s blood, the test was satisfied. The Court ruled, however, that the test was not dispositive: “Regardless, in stating that the ‘machine-or-transformation’ test is an ‘important and useful clue’ to patentability, we have neither said nor implied that the test trumps the ‘law of nature’ exclusion. That being so, the test fails here.” Mayo, at 19.

In light of these and other difficulties, it is likely that the Mayo decision will signal a general bias toward avoiding patents that claim natural laws as such, rather than establishing specific legal standards. Indeed, the Court expressly avoided stating exactly how much unconventional and non-routine activity was required to avoid claiming a law of nature itself. Mayo, at 18. That omission leaves the Federal Circuit a great deal of leeway in fashioning a more workable rule in later cases.

1 comment:

tim said...

How would you frame a claim to the law of universal gravitation? When you have done so, we can judge if a special exclusion of 'laws of Nature' is necessary to prohibit your claim.

'Processes' are patentable subject-matter under 35USC101. But not any process, surely? What was the Prometheus invention? " A method of optimizing therapeutic efficacy ..." And what did it consist of? A step of dosing a patient with a known drug, and a step of measuring the concentration of the drug in the dosed patient, qualified by a 'wherein' clause that 'indicates a need'. Is that a third step? - I don't think so. As far as I can see from a cursory glance at the prosecution file (don't take my word for this, check it!), both the first two steps were known, and known in combination. What wasn't known was that a particular result in step 2 'indicated a need'. You can't have a patent (surely) for doing what's been done before for a different reason (35USC102)? I think this decision could (just) be interpreted to make some sense of the 'law of nature' exclusion (which is prima facie unnecessary). But it's not going to be easy.