Ariosa Diagnostics v. Sequenom: Another One Bites The Dust

In Ariosa Diagnostics, Inc. v. Sequenom, Inc., 2013 U.S. Dist. LEXUS 156554, the U.S. District Ct. for N.D. Cal., granted summary judgment to Ariosa that the claims of U.S. Patent No. 6,258,540 were invalid as directed to patent-ineligible natural phenomena in combination with conventional sampling and detection techniques. The claim were directed to methods for detecting paternally inherited nucleic acid of fetal origin by amplifying a paternally inherited nucleic acid of fetal origin from a serum or plasma sample from a pregnant female and detecting the presence of this type of “cffDNA”. Claims 21 and25 recite that they are directed to “A method of performing a prenatal diagnosis” by obtaining a non-cellular fraction of the blood sample, amplifying the cffDNA in the fraction, detecting its presence and “providing a diagnosis based on the presence, quantity and/ or sequence of the DNA.”  Some conditions can be diagnosed by just measuring the absolute level of the bffDNA.  No specific diagnostic conclusions were recited in the claims. In fact, during oral argument, Sequenom had admitted: “I don’t disagree that if you go through all the elements in the claim you could put a check as either a conventional item or a natural phenomenon.”

The court reviewed all the “classic” case law cited in Myriad and Ariad relating to patent-eligible subject matter and found for Ariosa, “because paternally inherited bffDNA is a natural phenomenon and the claims of the ‘540 patent merely add well-understood, routine, conventional activity in the field to that natural phenomenon.”  (It helped Ariosa that the parties agreed that neither cffDNA nor its discovery in maternal plasma or serum is per se patentable since they are natural phenomena.) The narrower claims that recited more precise detection outcomes, e.g., requiring that the cffDNA is at least a certain percentage of the total DNA, were dismissed as merely limiting the natural phenomenon to specific types of that phenomenon. With respect to the claims that recited providing a diagnosis, the court simply stated that “Ariosa has presented…evidence that… providing a diagnosis based on fetal DNA were well-understood, routine, conventional activity, etc.”

I am left trying to puzzle out just what Sequenom was arguing supported its claims. Was it that it was inventive to obtain diagnoses based on measuring aspects of cffDNA? The court states that Sequenom argued that it was claiming a patent-eligible “use” of the DNA. However, Sequenom either did not argue, or did not convince the court that the claims were directed to an inventive “new use of an old compound” [my phrase]: “It is only an innovative or inventive use of a natural phenomenon that is afforded patent protection.. See Myriad, 133 S. Ct. at 2119… Thus, the only inventive concept contained in the patent is the discovery of cffDNA, which is not patentable.”

Sequenom also argued that, even though the presence of cffDNA in maternal blood is a natural phenomenon,  no one had used maternal blood as a source of cffDNA – in effect, Sequenom was arguing that this was the inventive concept in the claimed methods. The court’s reasoning in dismissing this argument gets cloudy here. On one hand, the court holds that “use of a newly discovered phenomenon… will not render a claim patentable if the use of that natural phenomenon…is the only innovation contained in the patent… [citing Flook and Mayo]… The court similarly concludes that paternally inherited cffDNA is not patentable simply because the claims contain steps indicating that it may be detected using existing DNA detection methods.”

Footnote 8 discusses Sequenom’s argument that the Supreme Court in Myriad “implicitly approved of claim 21” of US Pat No 5753441. This claim is drawn to a method for detecting a germline alteration in a BRCA1 gene by hybridizing a BRCA1 gene probe to an allele of one of a defined group of alterations and detecting the presence of a hybridization product. The Court just noted that the Supreme Court did not refer to any method claims in its opinion. (The Fed. Cir. also did not consider claim 21, but invalidated method claim 1 of that patent as an abstract idea.

As a final nail in the legal coffin that these claims were buried in, the Court held that the patent claims preempted the law of nature because there were no “commercially viable” alternative methods of detecting patenteral cffDNA. R.I.P ‘540 patent, which has also been attacked by Natera, Inc. and Verinata Health, Inc.

The question I am constantly asked these days is if diagnostic methods are still patent-eligible. Mayo is not well-characterized as involving diagnostic claims, at least Prometheus did not use a naturally occurring biomarker to diagnose any particular disease or condition. Intema presented diagnostic claims, but the biomarkers used and the methods of detecting them were well-known. Also the Fed. Cir. decision was labeled non-precedential. Intema petitioned for cert., arguing that its timed sampling method was an “inventive concept” but the argument that two samples/assays are better than one did not sound very inventive. Ariosa v. Sequenom involved claims to detecting a natural product using known techniques that also did not recite the step of providing a specific diagnostic conclusion.

So where are we in late 2013? If you are attempting to claim a diagnostic assay based on detecting a marker that is a natural product or a natural phenomenon, at least recite that the assay results in a specific diagnostic conclusion. Still, this is just the sort of claim that Justice Breyer criticized in the “Metabolite dissent”, which involved measuring the hCys level in a biological sample, where an elevated level of hCys indicated a cobalamin deficiency. I call this a simple “If a/then b” assay, e.g., if a man’s blood PSA level is determined to be above 3.5, concluding that there is an increased risk of prostate cancer. However, blood levels of hCys and PSA would probably be found to be natural phenomena, so patentee would be starting out in the hole. At the least, the legal community deserves a clear statement from the Fed. Cir. as to whether or not such a claim is still patent-eligible.

The next “level” of claim might be diagnostic assay that involves detecting a naturally occurring biomarker with a probe or capture moiety that application has invented. This recalls the years of patenting ELISA assays, and I still see and write such claims once in a while. If the inventor is the first to isolate the antigen as a pure chemical entity, e.g. a binding protein for a receptor on a lung cancer cell, and uses it to detect lung cancer, this assay should be patent-eligible. And, until a court rules otherwise, an isolated population of monoclonal antibodies are still patent-eligible. Let’s consider this as a guideline for 2014:

“If you can obtain a claim on any element of your method of diagnosis or detection claim, the claim is patent-eligible.”

For example, if you can patent the biomarker as a discrete chemical entity (and it is not a piece of genomic DNA), and/or if you can patent the detection moiety as a discrete chemical entity, the resulting assay/detection claim should be patentable. This follows from the principle that any use of a patented chemical is itself patentable. (That’s why we can re-join method claims in some cases.) However, to patent the PSA assay, the inventor would have to either be able to patent isolated PSA itself, or to patent the moiety used to detect it and/or the sampling and/or measurement techniques.   However, if your sampling or detection technique is patentable apart from using patent-eligible targets and/or binding moieties, this should be a sufficient “inventive concept” to get you past s. 101. This analysis is “cold comfort” for Intema or Sequenom, but I am still curious about Myraid’s “claim 21.”

The question I am most often asked today is “Can I patent this diagnostic assay?”

2 Responses to “Ariosa Diagnostics v. Sequenom: Another One Bites The Dust”

  1. David says:

    Great summary, thanks Warren. So it appears patents to diagnostic assays that rely on a “novel” biomarker and corresponding correlation fail to meet 101 threshold, if for example you use convention PCR or sequencing to assess the marker. Would you agree?

  2. Warren Woessner says:

    I think if you can “extract” and independently patent any aspoect of the assay – including the biomarker, the detection/quatification steps apart from the diagnosis – or any composition or device used to detect/quantify the biomarker (old or new), you can patent the diagnostic assay. In the US, any use of a “new compound” is patentable (so long as that compound is not genomic DNA or determined to be a patent-ineligible natural product)
    Warren