The broad language of § 101 of the Patent Act extends to “‘anything under the sun that is made by man” and excludes only laws of nature, natural phenomena, and abstract ideas. Diamond v. Chakrabarty, 447 U.S. 303, 309 (1980) (citation omitted). The patents-in-suit are not abstract, and there is nothing “natural” about them-except in the sense that everything under the sun is, ultimately, a product of nature. These patents describe specific, concrete processes for adjusting and optimizing the treatment of patients with particular man-made drugs, which metabolize into particular pharmacologically active compounds (e.g., 6-TG and 6-MMP) to achieve life-saving therapeutic efficacy while avoiding toxic side-effects. Such method of treatment claims accompany almost all patents on artificial drugs, and have routinely been enforced by this Court. To our knowledge no one, until this case, has ever suggested that such methods constitute unpatentable “natural phenomena” merely because the human body's reaction to an artificial substance is governed by natural laws. The same could be said of the foundational patents of the industrial age, like the Goodyear process for vulcanizing natural rubber or the Bessemer process for transforming molten iron into steel.

This Court and the Supreme Court have articulated a variety of approaches to the “natural phenomenon” question under § 101, including most recently the “machine-or-transformation” test adopted by this Court en banc in Bilski. The patents-in-suit pass that test with flying colors. The whole point of these processes is to transform the patient's body from a life-threatening physical condition into a healthier one. Along the way, the patient's body is transformed by administration of a synthetic thiopurine drug, producing metabolites never seen in nature. A sample of bodily fluid or tissue is transformed, with the help of sophisticated laboratory machines, to permit measurement of the levels of those metabolites. And the resulting data is transformed into a warning to the physician about the possible need to increase or decrease the patient's dosage.

The district court disregarded all of that because it believed, pre- Bilski, that a transformation of matter or data establishes patentability only for “industrial” processes. Rather than examining the purposes and effects of these processes as a whole, the district court dissected the patents into distinct steps and found reasons to disregard most of them until nothing was left but a “correlation” that it could describe (wrongly) as a natural phenomenon. The district court drew that approach to § 101 from the Supreme Court's decision in Flook. But as this Court has recognized several times, including in Bilski, the Supreme Court squarely rejected Flook's analysis in Diehr.⁵ Many patentable processes can be dissected into steps that, standing alone, would not be patentable.

In any event, the district court's analysis must be rejected even on its own terms. The metabolic consequences of administering a man-made drug are not a “natural” phenomenon but an artificial one. Of course those effects are mediated by natural laws, but so is combustion inside an automobile engine. Nonetheless, an improved method for tuning a car is patentable. The district court's suggestion that the “administering” and “determining” steps of the patents-in-suit can be disregarded as “merely necessary data-gathering steps for any use of the correlations, A00029, also betrays its complete failure to appreciate the purpose of these processes. Thiopurine drugs are not administered, and tissue samples are not obtained and transformed, merely to gather data for an abstract calculation. These steps are part of the ongoing treatment of desperately ill these processes. Thiopurine drugs are not administered, and tissue samples are not obtained and transformed, merely to gather data for an abstract calculation. These steps are part of the ongoing treatment of desperately ill patients. Presumably the district court would have dissected the patent in Diehr into data gathering steps (monitoring the temperature and pressure inside a rubber mold), a “natural” correlation (the prior art Arrhenius equation), and the “mental step” or “insubstantial post-solution activity” of sending a warning that it was time to open the mold. But the Supreme Court recognized the patent in Diehr for what it was: an improved method for curing rubber.

The district court's analysis would, if endorsed by this Court, threaten to invalidate the entire field of medical treatment and diagnostic patents on which the innovative and lifesaving biotechnology industry is largely built. It would also crush in its infancy the promise of personalized and genomic medicine, which of course will depend largely on insights into how man-made therapies interact with a particular patient's “natural” genes and body chemistry. There is absolutely no evidence that Congress intended to disincentivize such concrete, life-saving innovations in medicine and the life sciences. And there is no justification for reaching such a misguided policy result by needlessly expanding the narrow, judicially-created exclusion that denies patentability to abstract ideas and phenomena of nature.

The district court's summary judgment order should be reversed, and this case should be remanded with instructions to enter summary judgment under § 101 to Prometheus.

Prometheus fails to defend the claims under Section 101 as they were drafted and construed at its request. It repeatedly bends the claims, ignoring its own requested and gained construction, and instead suggests that the district court's requirement that someone “be warned” by the metabolite levels requires some real action. This so-called “warning,” however, requires nothing more than that a physician or other person recognize the correlation, i.e., that person is “warned” by the metabolite levels themselves of a possible needed change in dosage. And Prometheus repeatedly asserts that the claims involve patient treatment, when they in fact require a physician to do nothing. When Prometheus' wishful rewriting of the claims is stripped of these artifices, the claims are, like the claims in Labcorp, not even close to the dividing line drawn by Section 101.

On the merits, the claims do not satisfy Section 101 under any standard applied by this Court or the Supreme Court. The claims do not recite or depend on any particular machine or apparatus – the best Prometheus can do is suggest that measuring metabolite levels would involve a machine having certain generic components like pumps, detectors, and injectors. [Prometheus Br. at 31-32.] Nor do the claimed inventions physically transform subject matter – nothing happens in the claimed methods after the physician makes the correlation, and the transformations to which Pr
ometheus points are merely preparatory data-gathering. Finally, the claims fail the ultimate test of patentability by wholly preempting any practical use of the claimed correlation, because they are not limited in any way by what happens after the measurement is made – just like the claims in Flook and Labcorp. This Court should affirm.