In CareDx (Stanford U.) v. Natera Civ. Action No. 19-0567-CFC-CJB Consolidated (Sept. 28, 2021, D. Delaware), Judge Connolly ruled that the diagnostic method claimed in U. S. Pat. No. 8,703,652 and two others was a patent-ineligible natural phenomenon. The method is based upon the known correlation between the presence of donor-specific cell-free DNA (cfDNA) in a transplant recipient’s tissue and transplant rejection. Since the correlation was known, CareDx was left to argue that the improvement in sensitivity provided by the combination of known steps—which improvement was even recited in claim 1—was sufficient to satisfy the Mayo/Alice test for patent eligibility. However, by now we know that even discovery of a natural correlation and the discovery of its utility as a diagnostic is not enough to satisfy the requirement for the further “inventive concept” required by Step 2 of the Mayo/Alice test.

As almost a side-note, the Judge opined that the statement in Athena that Step 1 of the Mayo/Alice test could be satisfied by claims that recite only a natural law combined with “standard techniques for observing it”  both meet the “directed to” requirement of Step 1, and demonstrate the lack of any further “inventive concept” required for patent eligibility in Step 2 of the test. The judge cited case law asserting that “the two steps of the Alice inquiry overlap.” So if you fail Step 1 you instantly fail Step 2?

Although claim 1 is a page long, the steps of claim 1 were boiled down to four: 1. Obtaining a sample from the transplant recipient, 2. Genotyping the transplant donor and/or recipient to develop polymorphism or SAP “profiles”, 3. Sequencing the cfDNA from the sample(s) using multiplex or high-throughput sequencing or using digital PCR; and 4. Determining or quantifying the amount of donor cfDNA.

The Judge then went on to find the written description in the patent “is replete with characteristics of the techniques that confirm their conventionality.” See fn. 5 on pages 9-10. CareDx argued that “the challenged claims recite a series of specific, non-conventional laboratory techniques for detecting cell-free DNA with a high degree of sensitivity, in a manner that improves upon prior art methods of attempting such detection” [eds. Note: what happened to quantification?]. However the Judge framed the question:

“Because the asserted claims are directed to the detection of these natural phenomena, the dispositive inquiry under both steps of the Alice inquiry is whether the claimed methods are conventional (i.e., standard or routine). The patentee’s unequivocal and binding admission in the written description that the recited detection methods are conventional ends the matter before me.”

Oddly, the Therasense standards for inequitable conduct before the PTO are cited as support for a sort of reverse prosecution history estoppel by argument. To paraphrase, because CareDx made representations during prosecution that the steps of the claimed method are conventional, it cannot take the opposite position during litigation. But how is a competitor harmed by a patentee who later argues that a step described as routine is, in fact, unconventional?  Patentee is not arguing that any recited step is inoperable and so the claim is invalid under s. 112. Patentee is arguing that patentee had used early stage technology in a successful combination—even if they did not invent it.

Things go downhill from here. The judge accuses CareRx of falsely arguing that digital PCR is an “emerging technique” after he reads a paragraph from the specification describing the advantages of using digital PCR over prior PCR techniques. The judge also describes as “not true”, CareDx’s assertion that the specification states that the inventors “applied a never-before-used combination of techniques to better measure the correlation.” But even if the specification only broadly asserts that the claimed method is superior to prior art methods, should that matter if the argument about the novelty of the combination is, in fact, true?

I think that the Judge is just wrong when he disregards CareDx’s argument that “the novelty of the recited techniques to the detection of donor-specified [specific?] cfDNA makes the techniques nonconventional…Alice step two’s requirement of ‘additional features that must be new and useful’ is simply not met in this case because the asserted method claims recite standard detection techniques applied to naturally occurring phenomena…the asserted patents do not claim an ordered combination of the recited techniques[!]” The Judge himself characterized the claimed method has having 4 recited steps.

I am not a litigator but, if I were, I would not try so hard to argue that the individual steps are unconventional, and focus on the improvements due to the use of the steps in ordered combination. I would also have argued that the method resulted in the quantification of the cfDNA, not just its detection, to avoid the Ariosa buzz-saw. I think the Judge got irritated by what he felt were mischaracterizations of the specification and simply discounted the “ordered combination” when he finally got to it. Game, set, match.