« Infringing Ink | Main | Busy Day at the CAFC »

June 22, 2006

Patentability Unanswered

Labcorp's appeal to the Supreme Court (in Metabolite v. Labcorp) was closely watched by all involved in patent law, as it may well have decided the limits of patentability. In a peculiar hush, the Court brushed it aside; the dissent (SC 04-607) was deafening.

In short, Labcorp was found guilty of wiggling out of a license agreement on a patent claim, and found guilty of inducing infringement, over providing medical test results whereupon doctors would naturally understand a relationship to exist; the relationship being "a correlation between high levels of homocysteine in the blood and deficiencies of two essential vitamins, folate (folic acid) and cobalamin (vitamin B12)." The question (petition for certiorari) before the Supreme Court was whether patent infringement can occur when thinking about a relationship between two things, more specifically in this case, a correlation between test results and a medical condition.

The 15-page Supreme Court dissent provides an encapsulated history of the case [slightly edited for readability].

In the 1980s three university doctors, after conducting research into vitamin deficiencies, found a correlation between high levels of homocysteine in the blood and deficiencies of two essential vitamins, folate (folic acid) and cobalamin (vitamin B12). They also developed more accurate methods for testing body fluids for homocysteine, using gas chromatography and mass spectrometry. They published their findings in 1985. They obtained a patent. And that patent eventually found its commercial way into the hands of Competitive Technologies, Inc. (CTI), and its licensee Metabolite Laboratories, Inc. (Metabolite), the respondents here.

The patent contains several claims that cover the researchers’ new methods for testing homocysteine levels using gas chromatography and mass spectrometry. Supp. App. 30. In 1991, LabCorp (in fact, a corporate predecessor) took a license from Metabolite permitting it to use the tests described in the patent in return for 27.5% of related revenues. Their agreement permitted LabCorp to terminate the arrangement if “a more cost effective commercial alternative is available that does not infringe a valid and enforceable claim of” the patent. App. 305 (emphasis added). Until 1998, LabCorp used the patented tests and paid royalties. By that time, however, growing recognition that elevated homocysteine levels might predict risk of heart disease led to increased testing demand. Other companies began to produce alternative testing procedures. And LabCorp decided to use one of these other procedures - a test devised by Abbott Laboratories that LabCorp concluded was “far superior.”

LabCorp continued to pay royalties to respondents whenever it used the patented tests. But it concluded that Abbott’s test did not fall within the patent’s protective scope. And LabCorp consequently refused [in 1988] to pay royalties when it used the Abbott test. Id., at 237 (payment eliminated due to “change in methodology”).

In response, respondents brought this suit against LabCorp [in 1999] for patent infringement and breach of the license agreement. They did not claim that LabCorp’s use of the Abbott test infringed the patent’s claims describing methods for testing for homocysteine. Instead, respondents relied on a broader claim not limited to those tests, namely claim 13, the sole claim at issue here. That claim — set forth below in its entirety — seeks patent protection for:

“A method for detecting a deficiency of cobalamin or folate in warm-blooded animals comprising the steps of:
“assaying a body fluid for an elevated level of total homocysteine; and
“correlating an elevated level of total homocysteine in said body fluid with a deficiency of cobalamin or folate.”

Claim 13, respondents argued, created a protected monopoly over the process of “correlating” test results and potential vitamin deficiencies. The parties agreed that the words “assaying a body fluid” refer to the use of any test at all, whether patented or not patented, that determines whether a body fluid has an “elevated level of total homocysteine.” And at trial, the inventors testified that claim 13’s “correlating” step consists simply of a physician’s recognizing that a test that shows an elevated homocysteine level—by that very fact—shows the patient likely has a cobalamin or folate deficiency... They added that, because the natural relationship between homocysteine and vitamin deficiency was now well known, such “correlating” would occur automatically in the mind of any competent physician.

On this understanding of the claim, respondents argued, LabCorp was liable for inducing doctors to infringe. More specifically, LabCorp would conduct homocysteine tests and report the results... Doctors, because of their training, would know... a normal homocysteine range..., and would know that an elevated homocysteine level is correlated with a vitamin deficiency. Hence, in reviewing the test results, doctors would look at the... measure and automatically reach a conclusion about whether or not a person was suffering from a vitamin deficiency. Claim 13 therefore covered every homocysteine test that a doctor reviewed. And since LabCorp had advertised its tests and educated doctors about the correlation, LabCorp should be liable for actively inducing the doctors’ infringing acts. See 35 U.S.C. §271(b).

The jury found LabCorp liable on this theory. The District Court calculated damages based on unpaid royalties for some 350,000 homocysteine tests performed by LabCorp using the Abbott method. The court also enjoined LabCorp from performing “any homocysteine-only test, including, without limitation homocysteine-only tests via the Abbott method.”

LabCorp appealed. It argued to the Federal Circuit that the trial court was wrong to construe claim 13 so broadly that infringement took place “every time a physician does nothing more than look at a patient’s homocysteine level.”... Indeed, if so construed (rather than construed, say, to cover only patented tests), then claim 13 was “invalid for indefiniteness, lack of written description, non-enablement, anticipation, and obviousness.” LabCorp told the Federal Circuit:

“If the Court were to uphold this vague claim, anyone could obtain a patent on any scientific correlation—that there is a link between fact A and fact B—merely by drafting a patent claiming no more than ‘test for fact A and correlate with fact B’ . . . . Claim 13 does no more than that. If it is upheld, CTI would improperly gain a monopoly over a basic scientific fact rather than any novel invention of its own. The law is settled that no such claim should be allowed. See, e.g., Diamond v. Diehr, 450 U. S. 175 (1981); . . . Chisum on Patents §1.03[6].”

The Federal Circuit rejected LabCorp’s arguments. It agreed with the District Court that claim 13’s “correlating”step simply means “relating total homocysteine levels to cobalamin or folate deficiency, a deficiency in both, or a deficiency in neither.” That meaning, it said, is “discernable and clear”; it is definite, it is described in writing, and it would enable virtually anyone to follow the instruction it gives. And that is sufficient. The Court did not address LabCorp’s argument that, if so construed, claim 13 must be struck down as an improper effort to obtain patent protection for a law of nature.

Moreover, the Circuit concluded, because any competent doctor reviewing test results would automatically correlate those results with the presence or absence of a vitamin deficiency, virtually every doctor who ordered and read the tests was a direct infringer. And because LabCorp “publishes . . . Continuing Medical Education articles” and other pieces, which urge doctors to conduct the relevant tests and to reach a conclusion about whether a patient is suffering from a vitamin deficiency based upon the test results, LabCorp induces infringement. Finally, the court rejected LabCorp’s challenge to the injunction.

LabCorp filed a petition for certiorari. Question Three of the petition asks “[w]hether a method patent . . . directing a party simply to ‘correlate’ test results can validly claim a monopoly over a basic scientific relationship . . .such that any doctor necessarily infringes the patent merely by thinking about the relationship after looking at a test result.” Pet. for Cert. After calling for and receiving the views of the Solicitor General, we granted the petition, limited to Question Three.

In lower court, Competitive Technologies (the patent holder) was awarded $1.1 million in damages; Metabolite (the licensee, to which Labcorp was sublicensed) $4.5 million.

The S.C. agreed to take the case over a recommendation from the White House (Solicitor General) not to. 20 amici briefs piled in. Chief Justice John Roberts recused himself after taking part in early proceedings.

Oral arguments were heard before the Court in mid-March. Then, the tone of questioning foreshadowed rejecting the case, as the Justices showed concern that they may have accepted LabCorp's write of certori based on a defense that was never used in the lower courts, namely U.S.C. §101, covering the scope of patentability. In other words, the basis for hearing the case was murky, that in even taking the case, the S.C. had missteped. At that time, Justice Kennedy mused that it seemed imprudent to take a case based on subject matter not heard previously by the Court of Appeals (CAFC). The court dismissed the appeal without explanation; that may well have been the reason.

Of course, it's just as fair to say that LabCorp blew it by not setting up the patentability pins to be knocked down by the high Court. LabCorp should have framed the § 101 issue at the CAFC for appeal to the Supreme Court.

As to the merit of such broad patent protection, there was understandable skepticism. Perennial skeptic Justice Scalia snorted: "What was made by man here? When there's a presence of one substance in a human being, there is a deficiency of two others," adding that the claimed invention was merely the discovery of a natural principle.

Metabolite maintained that the claim went to making use of a discovery as part of a diagnostic step.

Today, three Justices dissented: Breyer, Stevens, and Souter, with simple logic: "The question presented is not unusually difficult. We have the authority to decide it. We said that we would do so." The dissent essentially agreed with Scalia's earlier musing that the claim was of a "natural phenomenon." The dissent also poked at the seminal 1998 State Street decision (p. 13) that opened the door to business method patents, which were derided in a concurring opinion in the eBay (v. MercExchange) case by the same Justices, along with Kennedy. At the same time, the dissent did note that the § 101 argument had not been considered by the CAFC, and that "we might benefit from [their] views."

Cases about §101, patentable subject matter, have rarely been put before the Supreme Court. In 1981, Diamond v. Diehr decided that a machine, using a computer, that transforms physical material, is patentable. In 1972, Gottschalk v. Benson ruled that a process for converting decimal numbers into computer binary (BCD, binary-coded decimal) was not patentable, as "merely a series of mathematical calculations or mental steps." In 1980, Diamond v. Chakrabarty decided that genetically modified microorganisms were patentable, while the 1948 Funk Brothers Seed Co. v. Kalo ruling declared soil innoculation not patentable as merely discovery of a law of nature. In the same area as the Labcorp case, the 1978 decision in Parker v. Flook ruled that "the discovery of... [a natural] phenomenon cannot support a patent unless there is some other inventive concept in its application."

Posted by Patent Hawk at June 22, 2006 12:07 PM | Litigation


Phillip Mann at IP Litigation Blog cut a classic entry about this - http://www.iplitigationblog.com/patent-lawsuit-news-89-we-dodged-a-bullet.html

Posted by: Patent Hawk at June 25, 2006 1:01 PM