In Roche Molecular Systems, Inc. v. Cepheid, the Federal Circuit affirmed the summary judgment decision of the U.S. District Court for the Northern District of California that held nucleotide primer claims and detection method claims invalid under 35 USC § 101. This decision may be at odds with USPTO guidance that claims directed to a method of detection without a diagnostic correlation are patent-eligible. Judge O’Malley’s concurrence highlights the problem of giving precedential effect to patent eligibility decisions rendered on limited factual records.

The Detection Method And Primers At Issue

The patent at issue was Roche’s U.S. Patent No. 5,643,723, which relates to primers and methods for detecting M. tuberculosis based on the detection of certain “signature” nucleotides for M. tuberculosis (MTB). According to the patent, the “invention provides a rapid, sensitive and specific process for detecting in vitro the presence of [MTB] and its drug-resistance phenotype.” In contrast, prior art methods required 3-8 weeks to confirm MTB infection and at least another 20 days to determine drug-resistance.

The Federal Circuit deemed claim 1 representative of the method claims. Claim 1 recites methods for detecting MTB based on the amplification of DNA containing signature nucleotides:

1. A method for detecting Mycobacterium tuberculosis in a biological sample suspected of containing M. tuberculosis comprising:
   (a) subjecting DNA from the biological sample to polymerase chain reaction using a plurality of primers…to amplify a portion of a M. tuberculosis rpoB gene…wherein the plurality of primers comprises at least one primer that hybridizes…at a site comprising at least one position-specific M. tuberculosis signature nucleotide selected…from the group consisting [of] a G at nucleotide position 2312, [+11 other sites]; and
   (b) detecting the presence or absence of an amplification product, wherein the presence of an amplification product is indicative of the presence of M. tuberculosis in the biological sample and wherein the absence of the amplification product is indicative of the absence of M. tuberculosis in the biological sample.

Dependent claims recite specific primers by SEQ ID NO.

The Federal Circuit deemed claim 17 representative of the “primer” claims. Claim 17 recites primers containing MTB signature nucleotides that could be used to distinguish MTB from other bacterial species:

17. A primer having 14-50 nucleotides that hybridizes under hybridizing conditions to an M. tuberculosis rpoB gene at a site comprising at least one position-specific M. tuberculosis signature nucleotide selected, with reference to FIG. 3 (SEQ ID NO: 1), from the group consisting of:
    a G at nucleotide position 2312, …[+11 other signature sites].

Again, dependent claims recite specific primers by SEQ ID NO.

The District Court Proceedings

The ‘723 patent was granted in 1997. This case originated in July 2014, when Roche sued Cepheid for alleged infringement based on Cepheid’s “Xpert® MTB/RIF Assay,” which is point of care assay for rapid (Inter Partes Review of claims 1-13 and 17-20 of the ’723 patent, which was denied. Cepheid then moved for summary judgement of invalidity based on lack of subject matter eligibility under § 101, which was granted.

The Federal Circuit Decision

The Federal Circuit decision was authored by Judge Reyna and joined by Judge Hughes and Judge O’Malley. Judge O’Malley also wrote a concurring opinion, which will be addressed in a separate article.

Bacterial Primers Are Just As Ineligible As Human Primers

Roche argued that the claimed primers should be found eligible at step one of a Mayo/Alice analysis, because they are different from naturally occurring DNA. In particular, Roche noted that the claimed primers are structurally different from naturally occurring MTB DNA because the MTB genome is circular, and so does not have a 3′-end or a free hydroxyl group at the 3′-end like a primer would.

However, the Federal Circuit agreed with the district court that this argument was “foreclose[d]” by the Federal Circuit decision in In Re BRACA1 (also known as Myriad v. Ambry). According to the Federal Circuit, the inquiry starts and ends with a comparison of the nucleotide sequences:*

It is undisputed that the primers before us have the identical nucleotide sequences as naturally occurring DNA, just like the primers found subject matter ineligible in BRCA1.

*****

Roche … contends that BRCA1 is distinguishable because, as a bacterium, MTB has “a circular chromosome, which has neither a 3-prime end nor a 3-prime hydroxyl [group],” while “[t]he primers at issue in BRCA1 were derived from human DNA, in which each chromosome occurs as a linear molecule.” Appellant Br. 23, 31.  Roche’s emphasis on the chromosome is misplaced. The shape of MTB’s chromosomes is not relevant to the inquiry on the subject matter eligibility of the claimed primers. As this court determined in BRCA1, the subject matter eligibility inquiry of primer claims hinges on comparing a claimed primer to its corresponding DNA segment on the chromosome—not the whole chromosome. …. Therefore, at Alice/Mayo step one, we find that the asserted primers are indistinguishable from naturally occurring DNA and that the primer claims are directed to a natural phenomenon.

*In footnote 5, the Federal Circuit left open the possibility that primers with altered DNA sequences or chemical modifications/labels “such that they cannot be isolated directly from naturally occurring DNA” might satisfy the patent eligibility requirement of § 101.
Gee, thanks.

Medically Important Primers Are Just As Ineligible As Other Primers

In what is becoming an unwelcome trope, the Federal Circuit acknowledged the significance of Roche’s discovery, but found it to be irrelevant, even under Alice/Mayo step two:

There is no doubt that Roche’s discovery of these signature nucleotides on the MTB rpoB gene and the designing of corresponding primers are valuable contributions to science and medicine allowing for faster detection of MTB in a biological sample and testing for rifampin resistance. However, “[g]roundbreaking, innovative, or even brilliant discovery does not by itself satisfy the § 101 inquiry.” Myriad, 569 U.S. at 591; Ariosa, 788 F.3d at 1379.

The primers are not patent-eligible because they can be found in nature, not because they are not valuable scientific discoveries.

Only New Laboratory Techniques Are Eligible For Patenting (?)

The Federal Circuit’s treatment of the method claims takes the “natural phenomenon” doctrine a step beyond the marker-diagnosis paradigm into the realm of pure detection methods.

At Alice/Mayo step one, the plain language of the asserted method claims, viewed in light of the written description, demonstrates that they are directed to naturally occurring phenomena. The method claims disclose a diagnostic test based on the observation that the presence of the eleven position-specific signature nucleotides of the naturally occurring MTB rpoB gene indicates the presence of MTB in a biological sample.*

*Note that the only “diagnosis” in claim 1 is determining the presence or absence of MTB in the biological sample.

Troublingly, the Federal Circuit’s analysis comments on the naturally occurring state of the biological sample and reminds us of their finding that the primers used in the methods are “indistinguishable” from naturally occurring segments of DNA. This is contrary to USPTO guidance to the effect that “the analysis of a process claim should focus on the active steps of the process rather than the products used in those steps,” which in turn was based on the Federal Circuit decision in CellzDirect.

Turning to Alice/Mayo step two, the Federal Circuit agreed with the district court that the application of “routine” PCR techniques to a “newly discovered natural phenomenon” did not amount to an “inventive concept” for patent eligibility. The Federal Circuit distinguished the methods at issue in CellzDirect:

Unlike the method claims of the ’723 patent, the invention in CellzDirect went beyond applying a known laboratory technique to a newly discovered natural phenomenon, and instead created an entirely new laboratory technique that “is not simply an observation or detection” based on the natural phenomenon. ….  In contrast, the ’723 patent claims a method of detection based on a natural phenomenon and employs only conventional, well-known laboratory techniques, which are the opposite of those at issue in CellzDirect.

In his remarks at the IPO Annual Meeting, Director Iancu emphasized that “Section 101 is about subject matter. It is meant to address categories of matter that are not ever eligible on their own, no matter how inventive or well-claimed they are.” How then is a method of subjecting a specific type of cells to repeated freeze/thaw cycles a patent-eligible laboratory technique, but a method of subjecting specific DNA sequences to an amplification process is not?

What Will The USPTO Do?

Claim 1 of Example 29 of the USPTO’s Life Sciences Subject Matter Eligibility Examples is directed to a detection method said to be patent-eligible:

1.  A method of detecting JUL-1 in a patient, said method comprising:
   a. obtaining a plasma sample from a human patient; and
   b. detecting whether JUL-1 is present in the plasma sample by contacting the plasma sample with an anti-JUL-1 antibody and detecting binding between JUL-1 and the antibody

The USPTO’s analysis finds that these method steps “do not recite or describe any recognized exception [to patent eligibility],” citing the Supreme Court decision in Mayo for the proposition that “steps of administering a drug to a patient and determining the resultant level of 6-thioguanine in the patient ‘are not themselves natural laws.'” Did the Federal Circuit forget about that Supreme Court guidance here?

While the USPTO is bound by Federal Circuit decisions, it likely will wait to see whether this decision is subject to rehearing or Supreme Court review before revising its guidance.