By Patent Publius
Earlier this month, the Federal Circuit issued its opinion in Ariosa v. Sequenom, a closely-watched biotechnology case with significant repercussions for patent-eligibility analysis generally. Unfortunately, the Federal Circuit misapplies the Supreme Court’s analytical framework from Mayo v. Prometheus, striking down Sequenom’s important innovation for the prenatal diagnosis of fetal abnormalities. The shame here is that the Mayo opinion itself was unnecessarily broad, and the Federal Circuit has now interpreted it to be even broader.
Section 101 of the Patent Act provides that “[w]hoever invents or discovers any new and useful process, machine, manufacture, or composition of matter . . . may obtain a patent therefor,” but there are judicial exceptions for “laws of nature, natural phenomenon, and abstract ideas.” Those exceptions are relevant here, where the Federal Circuit considers whether the claimed method of using cell-free fetal DNA (“cffDNA”) to make diagnoses is patentable subject matter.
In the Mayo opinion, the Supreme Court established a two-step analysis for determining whether method claims merely “set forth laws of nature” or instead apply those natural laws with “additional features” so as to become patent-eligible processes. The first step looks at whether the claims are directed to a patent-ineligible law of nature, and the second step looks at whether additional elements “transform the nature of the claim” into something that amounts to more than a claim on the law of nature itself.
Applying Mayo to the case at hand, the Federal Circuit’s analysis of the first step is perfunctory:
In this case, the asserted claims of the ‘540 patent are directed to a multistep method that starts with cffDNA taken from a sample of maternal plasma or serum—a naturally occurring non-cellular fetal DNA that circulates freely in the blood stream of a pregnant woman. . . . It is undisputed that the existence of cffDNA in maternal blood is a natural phenomenon. . . . The method ends with paternally inherited cffDNA, which is also a natural phenomenon. The method therefore begins and ends with a natural phenomenon. Thus, the claims are directed to matter that is naturally occurring.
The Federal Circuit’s conclusion that the method “begins and ends with a natural phenomenon” tells us very little of how this principle is to be applied generally. Certainly, the method begins with a biological sample of maternal plasma or serum that contains paternally-inherited cffDNA, and it makes sense to say that it begins with a natural phenomenon. Of course, everything begins with a natural phenomenon, so this is hardly instructive.
But it’s inaccurate to say that the method simply ends with cffDNA. The method itself takes the miniscule amount of cffDNA found in the sample and exponentially amplifies it to detectable levels. The resulting substance, unlike the beginning sample, gains significant and new utility from a diagnostic perspective. What comes out of the process is an artificially-enriched substance that, unlike the maternal plasma or serum fed into the process, can be used for many diagnostic purposes. That is, the method ends with a substance that is anything but a natural phenomenon.
Applying the second step of the Mayo framework, the Federal Circuit finds that Sequenom’s claimed methods are not significantly transformative:
Like the patentee in Mayo, Sequenom contends that the claimed methods are patent eligible applications of a natural phenomenon, specifically a method for detecting paternally inherited cffDNA. Using methods like PCR to amplify and detect cffDNA was well-understood, routine, and conventional activity in 1997. The method at issue here amounts to a general instruction to doctors to apply routine, conventional techniques when seeking to detect cffDNA. Because the method steps were well-understood, conventional and routine, the method of detecting paternally inherited cffDNA is not new and useful. The only subject matter new and useful as of the date of the application was the discovery of the presence of cffDNA in maternal plasma or serum.
The last sentence is the most perplexing: The “discovery of the presence of cffDNA in maternal plasma or serum” is what sets Sequenom’s method apart from that which was “well-understood, routine, and conventional activity in 1997.” The problem here stems from the Federal Circuit’s failure to consider the claimed method as a whole, as it purportedly sets out to do: “[W]e next consider the elements of each claim both individually and ‘as an ordered combination’ to determine whether additional elements ‘transform the nature of the claim’ into a patent-eligible application.”
Undoubtedly, some parts of Sequenom’s method were already well-known. No one denies, for example, that some of the techniques involved in amplifying and then detecting cffDNA were, in their general features, already conventional activity in the field (e.g., PCR). What makes the Sequenom method patentable is the sum of its parts, that is, the method as a whole that the Federal Circuit acknowledges to contain the new and useful discovery of cffDNA in the maternal plasma or serum.
This is the principal feature of Sequenom’s claimed invention and its central argument throughout the litigation. Yet, the Federal Circuit relegates it to one of “Sequenom’s remaining arguments” and addresses it in a brief paragraph near the end of the opinion, where it inexplicably claims: “This argument implies that the inventive concept lies in the discovery of cffDNA in plasma or serum. Even if so, this is not the invention claimed by the ’540 patent.” On the contrary, this discovery is anything but conventional, and the method as a whole transforms a natural phenomenon into something both artificial and patentable.
Overbroad (and Dangerous) Principles
The overbreadth of the Federal Circuit’s analysis threatens diagnostic methods across the board. If a method of detecting a natural phenomenon is always “directed to” that natural phenomenon, as the Federal Circuit suggests, then all such methods are prima facie patent-ineligible under the first step of the Mayo framework and must fight the uphill battle under its second step. This is particularly troubling since virtually all diagnostic tests detect natural phenomena. Moreover, the Federal Circuit’s application of the second step of the Mayo framework looks at each part of the method individually, ignoring the claimed method as a whole.
Not only is this principle breathtakingly broad in the damage it could cause to the diagnostics industry, it is neither required by, nor even consistent with, the controlling case law. Only claims to natural phenomena are per se patent-ineligible; however, applications of natural phenomena are generally patentable. Detecting a natural phenomenon is not the same thing as the phenomenon itself. It is instead a specific application of that phenomenon. While the Federal Circuit states that applications of natural phenomena are patent-eligible, it quickly proceeds to categorically suggest a principle under which all diagnostic inventions may have one foot in the Section 101 grave.
Another overly-broad principle from the Federal Circuit opinion comes from this statement: “For process claims that encompass natural phenomenon, the process steps are the additional features that must be new and useful.” This may at first seem obvious and uncontroversial, but in the context of the rest of the opinion, it proves quite problematic. The Federal Circuit cites Parker v. Flook as support: “The process itself, not merely the mathematical algorithm, must be new and useful.” But note the subtle distinction between the two quotes. The Supreme Court discussed the “process itself,” while the Federal Circuit discusses the “process steps.”
This distinction has two important effects. First, it is one of many signals in the opinion that demonstrates the Federal Circuit’s improper dissection of the claimed method into its components parts. Rather than consider whether the “process itself” is “new and useful,” as the Flook opinion had done, the Federal Circuit analyzes each step individually. There’s no consideration of how the steps integrate into the process as a whole, and there’s no mention of whether that entire process claims something other than the natural phenomenon itself.
Second, the Federal Circuit looks at each step in a very general way and ignores the details of the steps that confer patent eligibility. For example, the opinion spends much time discussing how routine the PCR method was at the time of filing. But Sequenom never claimed the PCR method itself. The Federal Circuit fails to address Sequenom’s central argument: The claimed method is a new process of detecting cffDNA by devising a novel sample source from which to extract it, namely, maternal plasma or serum. The application and adaptation of known techniques in this inventive way to a newly-discovered sample source is not conventional.
Finally, the most problematic and new principle that may emerge from this opinion is a subtle, yet very significant, extension of Mayo to invalidate claims directed to routine and conventional applications of natural laws. Mayo teaches that the mere addition of what is purely routine and conventional at the time of filing cannot save a claim directed to a law of nature: “In particular, the steps in the claimed processes (apart from the natural laws themselves) involve well-understood, routine, conventional activity previously engaged in by researchers in the field.”
The Federal Circuit appears to exclude from the patent system a routine application of a law of nature, rather than, as Mayo requires, a law of nature to which merely routine activities have been appended. That is, if one skilled in the art could, after being informed of a newly-discovered law of nature, use routine skill to arrive at the claimed invention, then that claimed invention may be invalidated under the Federal Circuit’s reasoning.
This is contrary to Mayo, and it could conceivably invalidate huge swaths of meritorious inventions. Once the principles underlying a new method are known, application of those principles to devise that method will very often be obvious. The Supreme Court has been very consistent in saying that applications of laws of nature are patent-eligible, including those applications that would have been obvious in view of newly-discovered laws of nature. It is a subtle, but important, point to recognize that Mayo did not say the opposite, as the Federal Circuit now interprets it.
The Preemption Question
One potential bright spot in the Federal Circuit’s opinion is its treatment of preemption. Instead of being a test for patent eligibility, preemption is properly understood as being solely a policy underlying eligibility exclusions. It can at most serve as an after-the-fact check on whether an already-reached conclusion of eligibility is consistent with this policy. The Federal Circuit here mostly validates this position:
The Supreme Court has made clear that the principle of preemption is the basis for the judicial exceptions to patentability. Alice, 134 S. Ct at 2354 (“We have described the concern that drives this exclusionary principal as one of pre-emption”). For this reason, questions on preemption are inherent in and resolved by the § 101 analysis. . . . Where a patent’s claims are deemed only to disclose patent ineligible subject matter under the Mayo framework, as they are in this case, preemption concerns are fully addressed and made moot.
This may ultimately be a hollow victory, however. The Federal Circuit also says: “While preemption may signal patent ineligible subject matter, the absence of complete preemption does not demonstrate patent eligibility.” The problem here is that it is impossible to ever show complete preemption because it is impossible to know at the time of filing whether something outside the claims could also be conceived. Inventions are, by definition, unforeseeable.
Moreover, allowing anything less than complete preemption to be sufficient to invalidate a claim threatens to invalidate far too much subject matter. By their very nature, patents are preemptive. Allowing courts and patent examiners to freely draw the line between allowable and prohibited levels of preemption invites unpredictable and arbitrary decisions based on personal value judgments. That very problem arose here, where the district court held the claims invalid, at least in part, because they covered what the judge deemed to be “the only commercially viable way of detecting” the embodiment of the law of nature.
The Promising Potential in Judge Linn’s Concurrence
Judge Linn’s concurrence is promising, but it falls short of its full potential. Judge Linn does a better job than the majority in recognizing and understanding the legal significance of the important facts of this case:
[N]o one was amplifying and detecting paternally-inherited cffDNA using the plasma or serum of pregnant mothers. Indeed, the maternal plasma used to be “routinely discarded,” . . . because, as Dr. Evans testified, “nobody thought that fetal cell-free DNA would be present.”
It is encouraging to see that a Federal Circuit judge has finally gone on record to point out the problems caused by ever-broadening applications of Mayo:
I join the court’s opinion invalidating the claims of the ‘540 patent only because I am bound by the sweeping language of the test set out in Mayo Collaborative Services v. Prometheus Laboratories, Inc. . . . In my view, the breadth of the second part of the test was unnecessary to the decision reached in Mayo. This case represents the consequence—perhaps unintended—of that broad language in excluding a meritorious invention from the patent protection it deserves and should have been entitled to retain.
Judge Linn errs, however, in his acquiescence that Mayo requires the majority’s conclusion. Judge Linn’s concurrence generally reads more like a dissent, but he undercuts his own criticism of Mayo and its effects by calling his opinion a “concurrence.” As he laments:
The Supreme Court’s blanket dismissal of conventional post-solution steps leaves no room to distinguish Mayo from this case, even though here no one was amplifying and detecting paternally-inherited cffDNA using the plasma or serum of pregnant mothers.
But the second half of this sentence shows the critical distinction that makes Sequenom’s claims patent-eligible, even in view of Mayo. Unlike the claims analyzed in Mayo, Sequenom’s process is new and not routinely engaged in by researchers in the field. Judge Linn even states the point better elsewhere in his own concurrence:
Unlike in Mayo, the ‘540 patent claims a new method that should be patent eligible. While the instructions in the claims at issue in Mayo had been widely used by doctors—they had been measuring metabolites and recalculating dosages based on toxicity/inefficacy limits for years—here, the amplification and detection of cffDNA had never before been done.
Judge Linn should be praised for critiquing Mayo as bad law that has led to the invalidation of untold meritorious patent claims. Unfortunately, however, he may have unintentionally contributed to the expansive scope of Mayo about which he complains by failing to factually distinguish (and hence cabin) the Supreme Court’s opinion when presented with such a good opportunity to do so.
All told, the Federal Circuit’s opinion in Ariosa v. Sequenom is a predictable, yet unfortunate, application of the Supreme Court’s disastrous reasoning in Mayo. The unintended consequences of the Supreme Court’s opinion have been further realized in the Federal Circuit’s denial of Sequenom’s innovative claimed method for diagnosing fetal abnormalities. Only time will tell how many other innovations will suffer under the Supreme Court’s careless expansion of Section 101’s patent eligibility analysis.