Sequenom Files Opening Brief in Appeal of Summary Judgment on Section 101 Grounds

By Kevin E. Noonan --

Earlier this week, Sequenom, Inc. filed its opening brief in Ariosa Diagnostics, Inc. v. Sequenom, Inc., appealing summary judgment that its licensed claims to a genetic diagnostic method for detecting fetal diseases and abnormalities in utero were invalid for reciting patent-ineligible subject matter.  In view of the exquisitely subjective nature of the standards for patent eligibility enunciated by the Supreme Court (most recently in Mayo v. Prometheus), Sequenom may face a task more daunting that it should be.  But in view of the fact that anything short of reversing the District Court's decision below can be expected to reduce the likelihood that personalized medicine will become commercially viable in our lifetimes, it is important that they prevail.

To recap, the technology at issue was non-invasive prenatal diagnosis of sex determination, blood typing, other genetic disorders and detection of pre-eclampsia, using a simple blood test that reduces or eliminates the need for amniocentesis and chorionic villus sampling (which incur risks to both mother and child).  Sequenom Inc. is the exclusive licensee of U.S. Patent No. 6,258,540 originally obtained by Isis Inc.; there were three independent claims attacked by Ariosa and asserted by Sequenom in its patent infringement counterclaims:

1. A method for detecting a paternally inherited nucleic acid of fetal origin performed on a maternal serum or plasma sample from a pregnant female, which method comprises
 amplifying a paternally inherited nucleic acid from the serum or plasma sample and
 detecting the presence of a paternally inherited nucleic acid of fetal origin in the sample.

24. A method for detecting a paternally inherited nucleic acid on a maternal blood sample, which method comprises:
  removing all or substantially all nucleated and anucleated cell populations from the blood sample,
 amplifying a paternally inherited nucleic acid from the remaining fluid and subjecting the amplified nucleic acid to a test for the Paternally [sic] inherited fetal nucleic acid.

25. A method for performing a prenatal diagnosis on a maternal blood sample, which method comprises
 obtaining a non-cellular fraction of the blood sample
 amplifying a paternally inherited nucleic acid from the non-cellular fraction and
 performing nucleic acid analysis on the amplified nucleic acid to detect paternally inherited fetal nucleic acid.

The District Court based its decision on the Supreme Court's decision in AMP v. Myriad Genetics, that "naturally occurring" DNA is not patent eligible as a "product of nature," coupled that with the proscription from Mayo that "merely" adding "routine, conventional and well-understood" process steps to a "natural law" (or, in this case, "product of nature") was not "enough" to render method claims patent-eligible, determined that the amplification and sequencing steps recited in the claims were "routine, conventional and well-understood" process steps, and finally invoked Parker v. Flook for the proposition that the Supreme Court did not recognize any distinctions between an "abstract idea" and natural laws or phenomena to arrive at its decision that the claims at issue did not satisfy Section 101.  The District Court also discounted evidence that the art had developed alternative, non-infringing methods for performing the claims detection and diagnosis methods, on the grounds that these methods were not contemporaneous and were not "commercially viable."  The District Court entered summary judgment last November, and Sequenom timely appealed.

Sequenom presents its arguments in thre parts:  first, that the Federal Circuit should review the District Court's determination de novo, and should apply the exceptions to subject matter eligibility narrowly while reading Section 101 expansively, consistent with Congressional intent.  Second, the Court should recognize that the District Court erred in how it applied controlling precedent to the claims at issue, first because the claims do not "merely" claims a "natural phenomenon" and second that the lower court applied a standard (i.e, the "commercially viable" standard for noninfringing alternatives that "lacks legal basics and contradicts public policy").  And finally, whether the District Court erred by deconstructing the claims and assessing their patent-eligibility separately, contrary to the Supreme Court's teachings in Diamond v. Diehr (but, honestly, consistent with Justice Breyer's treatment of the claims in Mayo, a case that pointedly did not overrule Diehr).

The brief first begins with a section on the nature of the invention, which points our to the Court the benefits to the public of the invention disclosed and claimed in the patent-in-suit.  These include the ability to detect fetal abnormalities earlier in pregnancy and without the need for invasive procedures like amniocentesis and chorionic villus sampling that can put both mother and child's lives at risk.  The facts elucidated include those that support the novelty and non-obviousness of the claimed methods, factors necessitated by the Supreme Court's focus on whether a claim is "inventive" in Mayo (thus turning back the clock to the state of the law before Congress affirmatively created Section 103 in the Patent Act to inhibit if not prevent this type of subjective determination of what was patent-worthy by the Court).  (In homage to this pre-1952 Act Supreme Court precedent, Sequenom characterizes the inventors' recognition that cffDNA could be found in the acellular fraction of maternal blood as an "original stroke of genius" and note that the scientific journal article announcing this discovery had been cited "over a thousand times.")  And bowing to more recent trends from the Court, Sequenom accentuates in its factual allegations that the claimed methods were "limited" to detecting paternally inherited DNA in the cffDNA in maternal blood.  Finally, the fact statement notes that the claimed methods "transform" the cffDNA in application of the process steps, including fractionation, amplification and detection (including sequence determination) of the amplified cffDNA.  This section also sets forth the alternative methods for detecting cffDNA and using said detection for diagnostic applications.

The brief also reminds the Court that, because this was a decision on summary judgment (under Ninth Circuit precedent) and because subject matter eligibility is a question of law, the Court is entitled and should decide the question before it de novo.  In so doing, Sequenom urges the Court to apply patent eligibility broadly and the exceptions to it narrowly, based on the statutory language and the Supreme Court's construction of the scope of the statute in Diehr and Bilski v. Kappos as well as recent Federal Circuit precedent, such as CLS Bank v. Alice and Ultramercial v. Hulu (perhaps in an effort to show that, no matter how the cases before these courts have come out, the principle of expansive reading on the scope of Section 101 is consistently broad).  The brief cites much of the same precedent for the corresponding principle that judicially created exceptions to broad subject matter eligibility should be applied parsimoniously.  In this context, of course, Ariosa had the burden of proving that the exception should be applied against Sequenom's claims by clear and convincing evidence, citing CLS Bank, and while not affirmatively drawing the conclusion suggests that Ariosa failed to meet this burden.

In its first argument, Sequenom asserts that its claims do not claim a natural product or phenomenon, nor preempt all uses of fetal DNA or cffDNA, but rather claim narrow, specific, and limited applications of the biological act that cffDNA is present in maternal blood.  The brief identifies undue preemption (defined as "all other uses of the phenomenon"), which Sequenom argues is not the case here due to the existence of alterative, non-infringing methods for detecting cffDNA (once these inventors had disclosed that it was present in maternal blood).  Sequenom argues that the District Court erred by assessing subject matter eligibility of its claims as if they claimed the cffDNA itself, rather than an application of methods for detecting such DNA and providing a diagnosis of fetal disease or developmental disorder therefrom.  Sequenom also argues that the existence of the identified alternative methods for detecting cffDNA was dispositive on the question of preemption, and further that the claims taken as a whole satisfied the statutory requirements of novelty and non-obviousness as well as the Supreme Court's subjective "inventive concept" test and thus are patent eligible.

Regarding preemption, the brief argues that the Supreme Court's concern with preemption was that a claim "preempts all practical uses" of a natural law, phenomenon of nature, or abstract idea, again citing Supreme Court (Mayo, O'Reilly v. Morse) as well as Federal Circuit case law.  This primacy for the preemption question was ignored by the District Court, according to Sequenom's brief, which deemed the preemption issue to be only "a consideration" when deciding whether a claim satisfies Section 101 or falls within an exception to it.  According to the brief:

The Supreme Court has variously stated the distinction it drew in the Morse and Telephone cases, but the essence is that a patent recites ineligible subject matter only when it claims for itself, or preempts all other uses of, an abstract idea, law of nature, or natural phenomenon.

The brief then contrasts this proscription of patent eligibility with instances where "a patent applies a natural phenomenon in a limited, non-preemptive manner so that other uses may be made of it," citing Bilski and Diehr (emphasis in original).  In this regard the brief argues that the District Court misread Flook, Bilski, and Mayo to be directed to "non-preemptive" patent claims, and thus failed to recognize the importance of the preemption question, and that the failure to unduly preempt prevented Sequenom's claims from invalidity under Section 101.  And the brief identifies Diehr as the case most closely related to the facts in this case (and cites how the Diehr Court distinguished the claims before it from the Flook case), asserting error by the District Court in "giving short shrift" to the Diehr decision.  Thus:

In sum, the Supreme Court's and this Court's precedents follow a consistent theme:  where claims recite a natural phenomenon and no more, or when they recite a method in terms so general that it covers all ways to use the natural phenomenon, then the claims are not patent-eligible.  Where, as here, the patent claims a specific limited method and there are alternative methods available, then there is no preemption and no Section 101 eligibility concern.  The District Court failed to follow this controlling Section 101 law.

With regard to the second argument, Sequenom asserts that the District Court "discounted" its evidence for alternative, noninfringing methods, and that "[t]he District Court adopted an unprecedented and improper standard to determine the relevance of alternative methods offered to prove the claimed method does not preempt all uses of a natural phenomenon.  It held that an alternative method would be relevant to show lack of preemption only if it both (i) was disclosed publicly before the '540 patent issued, and (ii) was shown by Sequenom to be commercially viable, a standard not enunciated by either the Supreme Court nor the Federal Circuit.  Further, the brief argues that there is no "rule or logic" in the District Court's "commercially viable" standard for alternative methods for practicing the claimed methods.  The brief notes that the concern has always been that claims "tie up future invention" and thus strongly suggest both that Sequenom's evidence for after-arising non-infringing alternatives was relevant and that this evidence supported Sequenom's arguments for subject matter eligibility.  Sequenom's claims do not do so and thus the District Court erred according to the brief.  Sequenom sets forth the consequences of the misapplied thinking of the District Court:

As the Supreme Court observed in Myriad, it is expected that the original discoverers of a natural phenomenon or law of nature will invent the first (and perhaps best) method of applying their discovery.  . . .  However, the District Court's rule would require inventors, such as Lo and Wainscoat -- who discovered cffDNA in 1996 and claimed a method applying it a few months later -- to wait, perhaps indefinitely, to patent their method until other inventors have disclosed alternative methods using the discovery.

Further, the District Court's "previously disclosed" requirement turns the presumption of validity on its head.  Under the District Court's rule, the first-to-invent is branded as a "preempter" whose method is doomed to be patent-ineligible.  Because Lo and Wainscoat came up with the first-filed method applying their ground-breaking discovery of cffDNA before any alternatives had been published, under the District Court's reasoning, their patent necessarily preempts all other methods that could apply their discovery.  Someone has to file first, and the first inventors should not be required to hold back disclosure of their method until others disclose alternative methods.  The net effect of the waiting game the District Court's rule creates would be to stymie the disclosure and exploitation of inventions -- the reverse of the incentives the patent laws are intended to foster.

(citations omitted).

Specifically addressing the "commercially viable" standard enunciated by the District Court, the brief argues that "[t]he "commercially viable" requirement would impose a higher standard on alternative, non-preemptive methods than now exists for patented methods," and argues that this is a rule unsupported by legal precedent nor logic itself.  In making these arguments, the brief contrasts the District Court's standards for non-infringing alternatives with the requirements of patented methods, wherein there is no requirement that a claimed invention be commercially viable (citing Juicy Whip, Inc. v. Orange Bang, Inc., 185 F.3d 1364, 1366 (Fed. Cir. 1999), CFMT, Inc. v. YieldUp Int'l Corp., 349 F.3d 1333, 1338 (Fed. Cir. 2003), and Barmag Barmer Machinenfabrik AG v. Murata Machinery, Ltd., 731 F.2d 831, 839 (Fed. Circ. 1984) as examples).  This portion of the brief also notes that, under the District Court's rubric many earlier patents would need to have been invalidated due to lack of commercially viable, non-infringing alternatives.  (And the brief asserts that there is no definition or other factual basis or standard for what "commercially viable" requires or entails, while providing a propensity to invalidate the best method for a particular purpose.)

Finally, Sequenom argues that the District Court misinterpreted and misapplied the law regarding the need for an "inventive concept" in the claims.  Here the error Sequenom asserts is that the District Court did not consider the claims "as a whole" but rather cherry-picked the different limitations and evaluated them separately and independently for patentability.  The brief argues that this analytical style is contrary to Supreme Court precedent, and then argues that the term is not a cipher for novelty and non-obviousness but rather a way of assessing "genuine human contribution" to an invention, citing (and trying to apply) Supreme Court precedent on this question.  And even using the District Court's failed analytical tools these claims should be deemed "inventive," according to the brief, based on the conventional practices in the art (including discarding acellular blood fraction containing cffDNA).  On the subject of inventiveness the brief has this to say:

What is probative of patent-eligibility, and what the District Court erroneously denied, was the patent's combination of these steps for the first time in a groundbreaking method for the purpose of detecting paternally-inherited cffDNA in maternal plasma or serum and diagnosing fetal characteristics.  This new combination for this new purpose was neither conventional nor routine, and it satisfies the requirements of Section 101.

It is the failure to look at the combination "as a whole" that was the source of the District Court's error.

In the final section of the brief, Sequenom reminds the Court that this case was decided on remand from the Federal Circuit with instructions to consider the issues before the District Court in light of the Myriad decision, and that the District Court misapplied that law as it relates to patent-eligible method claims for screening anticancer drugs, analogizing the laboratory transformed variant with cffDNA in this case.  The brief reads the Court's Myriad decision as "[drawing] the patent-ineligibility line tightly around the genes' DNA sequences themselves" and that the Myriad decision "holds that even a small step away from the natural phenomenon as the result of human intervention is sufficient for patent-eligibility."  In the claims before the Court, Sequenom argues, the cffDNA is much more like cDNA deemed patent-eligible under Myriad than naturally occurring genomic DNA (which the Myriad Court deemed patent-ineligible).  And the brief argues that these claims fall squarely within the types of patent-eligible claims and methods cited with approval in Section III of the Myriad decision itself.

Patent Docs will provide additional posts on Ariosa's responsive brief as well as any amicus briefs filed on behalf of either party.

Comments

[EG]

Kevin,

I do wish Sequenom well in its effort to convince the Federal Circuit to render order out of this utter chaos that patent-eligibility under 35 USC 101 become. As I've stated before, the primarily responsibility for this mess lies at the foot of Our Judicial Mount Olympus who have so mucked up this area of patent law that no one (including them) knows how to apply this patent-eligibility precedent other than subjectively, irrationally, illogically and with no consistency.

[Fred]

Very nice article I appreciate the thoroughness. I think this is a very tricky arena. I am not an attorney but a scientist. I think '540 is a very weak patent from a methods point of view (and not from a Myriad point of view). First let's examine an empirical argument that, I believe, has not much legal merit but is still a valid argument from a rational scientific viewpoint. This patent is old from the point of view of DNA sequencing technology. If there really was a workable method in the patent based upon PCR (which was already an old technique at the time), how come it took so long for Sequenom and Lo to come up with a method that actually works? Instead they spent a lot of time on another method that fell outside the domain of '540 the so-called PLAC RNA method. This method doesn't work even though Sequenom and Dr. Lo said it did and it led to the fraud that destroyed Sequenom's market capitalization in 2009. If there really was a workable "method" in '540 then a) MathernT21 would have appeared 10 years ago and b) they wouldn't have needed the paternal DNA limitation. If I decide to patent consciousness on a chip, and 15 years later someone actually figures out how to do it can I then claim "Hey, I invented that". Unfortunately, the patent portfolio of the US is littered with patents like this - many of which end up getting enforced anyway.

In my opinion, Ilston's claim construction of the use of the word paternal, was thrown aside erroneously. The reason for the need to use paternal DNA was that, at the time, massively parallel DNA sequencing did not exist. Therefore, there was a real question as to how one would count the excess fetal DNA that would lead to a diagnosis of aneuploidy in the presence of the huge maternal background. This was a vexing scientific question and it is pretty obvious to any scientist with experience in this domaIn (though apparently not to judges) that '540 does not provide a method for how to do this. Therefore, in the original patent examination they were required to use the paternal limitation in claim 1 (and that is a limitation inherited in every subsequent claim in that patent) as that was the only means, at the time, that could readily distinguish between the maternal and fetal DNA. Fast forward a few years and Fan and Quake from Stanford put out a patent in '017 or "Non-invasive Fetal Genetic Screening by Digital Analysis" in which one of the claims is to use massively parallel sequencing to determine aneuploidies. Once one sees this it is now obvious how to go about counting the DNA to anyone skilled in the field. It is also obvious that one doesn't need the paternal limitation (although it is true that by counting both the fetal and maternal DNA that some of the DNA came from the father - DUH). Frankly, from an inventive scientific viewpoint the only real novelty in '540 is the idea of using cell free fetal DNA in maternal plasma. A great idea to be sure, but probably limited by Myriad. One still has to figure out how to implement that idea, and that is where '540 falls on its face. There is no way, at the time the patent was written, that one could use the very simple PCR methods therein to actually perform anueploidy analysis and do the very exacting counting that was required to have a statistically usable test. Every scientist I have talked to about this agrees, although it wouldn't be hard to find expert witnesses or amicae briefs that would suggest otherwise.

On another, more general, point you raised e,g, "But in view of the fact that anything short of reversing the District Court's decision below can be expected to reduce the likelihood that personalized medicine will become commercially viable in our lifetimes, it is important that they prevail." I would disagree. The large numbers of companies that entered into the NIPT space that is currently the subject of this article would suggest otherwise. It is clear that there is a lot of commercial value that will be pursued even with the current state of the patent environment. And I don't see any overly strong impediments to strong competition in this space of personalized medicine - especially given how much of the ip came from the Human Genome Project and how many non-profits (e.g. the J. Craig Venter Foundation) are engaged in this pursuit. But again, I am not a lawyer and I may have a naive viewpoint.

By the way, in the interests of full disclosure, I am long Sequenom stock, but I just wish they would stop spending massive amounts of money defending what I believe is a very weak patent.

[Kevin E. Noonan]

Dear Fred:

Interesting perspective. From the legal point of view, I would have no issue invalidating the '540 patent on any of the grounds you raise, which is the traditional way to do it: the patent claims are not novel, are obvious, are inoperable, are not enabled, etc. I have no desire to preserve patents that objectively fail the many tests for patentability.

The problem here is that the judge invalidated the claims categorically - "these types of claims (no matter how novel, useful and non-obvious) cannot be patented." And she did so based on her subjective view that these claims (and their ilk) should not be patented.

As for the prediction, we are in an interesting time - lots of DNA information from the HGP but less (not zero) information about how to predict diseases or adverse drug reactions based on genotypes. Yes, there are lots of people in the space but because of the complexity of the task disclosure will become more and more of a problem - without a way to recompense the costs of not discovery but commercialization there will be incentives not to disclose. This may or may not come to pass, but we know the world with patents - more than a million women having been tested for the BRCA gene mutations, and those methods become freely available in about 2 years. In a future without patents, it is much harder to predict what will happen.

Thanks for the comment.

[Skeptical]

Dr. Noonan,

I believe that Abraham Lincoln had something to say about that.

(as would Congressmen Bayh and Dole)

Perhaps the next President will give more than one liners and empty platitudes to innovation.

[6]

"But in view of the fact that anything short of reversing the District Court's decision below can be expected to reduce the likelihood that personalized medicine will become commercially viable in our lifetimes, it is important that they prevail."

You know Kev, if it is that important to the commercial viability of such then congress can always be lobbied for an exception to the exceptions. It isn't like they're immune to evidence of failure in the marketplace because of no patent protection if that should occur here shortly.

Though frankly I'm a bit wary of your alleged foreseeing the sky falling in this area.

[JNRoss]

Kevin: "in view of the fact that anything short of reversing the District Court's decision below can be expected to reduce the likelihood that personalized medicine will become commercially viable in our lifetimes, it is important that they prevail."

I don't think even Kevin believes this. But maybe he believes it's a good sound bite in defense of his clients' interests so he repeats something similarly ominous every time a case turns up that might affect those interests.

The fact is that various forms of "personalized medicine" have been with us for a long, long time and as more (unpatentable) information is made available to doctors, more "personalized" treatment can be provided to patients. Of course, there's also the issue of what insurance companies, who currently function as middle-men/parasites in our healthcare system, will do with the information.

"From the legal point of view, I would have no issue invalidating the '540 patent on any of the grounds you raise, which is the traditional way to do it: the patent claims are not novel, are obvious, are inoperable, are not enabled, etc."

Then why do you insist on burying such an admission in the comments? As an "expert" on this subject, your views would be welcomed by the public -- you know, vast majority of non-patent owning people whose well-being you are allegedly so concerned about. If you this patent is obvious, non-enabled junk then you should be spreading the news because, if that's the case, then everyone will be paying more for whatever test falls within the scope of the junk claims than they otherwise would be.

[JNRoss]

"In view of the exquisitely subjective nature of the standards for patent eligibility enunciated by the Supreme Court (most recently in Mayo v. Prometheus)"

There was nothing "subjective" about the decision to declare the claims at issue in Prometheus ineligible. You do understand that now, don't you, Kevin?

Mayo was using its claims to prevent practitioners of the prior art from thinking about an inelgible fact. You can't use patents to protect facts (or any other ineligible subject matter). There's nothing "subjective" about that, unless you believe that one should be able to patent facts. And if you believe that, you should just come right out and say it rather than pretending you are making some "principled" argument about the reasoning in Prometheus.

[JNRoss]

A friendly reminder about what you wrote, Kevin, last August in a comment thread about this case:

"I agree that the patent eligibility of the claim will depend on whether detecting fetal DNA in maternal blood was "new" when the patent application was filed."

Subsequent interesting discussion here last November after the Judge Illston's decision:

PRENATAL SEX DETERMINATION BY DNA AMPLIFICATION FROM MATERNAL PERIPHERAL BLOOD

That's fetal DNA detection in maternal blood, by one of the inventors, published in Lancet in 1989, 7 years before the patent at issue here was filed. This paper is mentioned in the background section of specification. Also this publication from the inventors in 1993:

http://link.springer.com/article/10.1007/BF00217445 "Prenatal sex determination from maternal peripheral blood using the polymerase chain reaction"

At the time of publication of those papers, it was already known that PCR would and could detect minute amounts of DNA. I haven't read these two papers to read all the details but at least one Ph.D. student wrote the following: "Lo et al. (1989) have demonstrated by PCR the existence of fetal cell-free DNA in maternal plasma ..." Perhaps that student misread the paper ...

Regardless, it boggles the mind that nobody in the field suggested checking whether some of that fetal DNA was cell-free around the time that those earlier papers were published. I would expect such an analysis would show up as a control in those papers circa 1989-1993. Very odd.

[J. Lightner]

Kevin,
Thank you for this informative post.
I am wondering to what extent the patentability of Sequenom's method bears an analogy to extractive mining, an area where I have greater expertise.
There are some common leaching methods to extract minerals and metals, such as gold, from mine tailings. Each method involves several engineering steps, which are in themselves complex and can be discussed in isolation, but are combined in well known ways for the process.
Would the application of one such common method specifically to extract a different mineral from a different waste stream merit a patent - if nobody had previously thought to apply the method to extract that different mineral or metal?
I don't know if there is a general answer to my question. However I would much appreciate your thoughts on the relevance of the analogy to the Sequenom case, if you have time.
Thank you

[Kevin E. Noonan]

Dear JN:

Thanks for supporting my position - there were ample ways to invalidate this claim under 102 instead of 101, and I think that would be the better way.

As for Prometheus, how they were trying to enforce the claim isn't the point - if you can find a practical standard that can be applied objectively (and not in the "we know it when we see it" sense) from Justice Breyer's decision, please let me know. As I see it, the lack of objective standard leads to a decision like the one in Ariosa.

As for the future of personalized medicine, we shall see - I have some faith in my fellow practitioners' ability to adapt to whatever nonsense the courts come up with.