Personalized Medicine and Patent Eligibility under 35 U.S.C. § 101

Double-digit growth is predicted for the molecular diagnostic market in the next several years, driven in part by an increased demand for personalized medicine—medical decisions and therapeutic products tailored to address the precise needs of each individual patient.[1]  To date, the FDA has approved 19 companion diagnostics, those assays that must be performed on a patient to ensure that the corresponding therapeutic will be safe and effective.[2] The agency has also approved other nucleic acid based tests useful for a personalized medicine approach to certain cancers, cystic fibrosis, and various metabolic or clotting disorders.[3]

Two recent Supreme Court cases interpreted patent eligibility under 35 U.S.C. § 101 and addressed the patent eligibility of genes and applications of laws of nature. Those cases will almost certainly impact the patentability of inventions important to personalized medicine practices. After the Court’s decision in Myriad Genetics, genes are not patent-eligible when claimed in reference to their genetic sequence as it exists in nature.[4] Furthermore, diagnostic assays or methods are not patent-eligible under the Court’s holding in Prometheus if those claims only articulate a naturally-occurring phenomenon or relationship.[5]  The key to both cases for patent holders is to claim something “new.” A claim might be directed to a non-naturally occurring form of a gene: for example, a gene claimed in a recombinant form. And claims to methods that use (or rely upon) genetic information should include steps that go beyond that which is “well-understood, routine and conventional activity already engaged in by the scientific community.”[6] The latter, at least, is more easily said than done, and patent holders will continue to probe the boundaries established by Myriad and Prometheus.

Two recent cases illustrate the challenges faced by litigants and courts alike.

Prenatal Testing

Sequenom lost a patent suit against Ariosa Diagnostics last fall. The companies are rivals in the prenatal diagnosis market, which also includes Illumina and Natera.[7] At issue was whether Ariosa’s Harmony Prenatal Test infringed Sequenom’s patent on non-invasive prenatal diagnosis.[8] The case is currently pending before the Federal Circuit.[9]

Sequenom’s patents cover a method for detecting paternally inherited nucleic acids of fetal origin from maternal serum samples.  The claimed method has two basic steps: amplification of paternally inherited nucleic acid, and detection of the presence of paternally inherited nucleic acids in a maternal sample. Other claims narrow the method, specifying that the amplification be performed by PCR, or by reciting a particular gene, type of gene, or chromosome.

The trial court heard the parties’ arguments concerning the patent eligibility of Sequenom’s patent. Ariosa argued that cell-free, fetal DNA (“cffDNA”) detected in its assay is naturally-occurring, and that other aspects of the claims reflect well understood, routine and conventional activities in the field.[10] Sequenom disagreed. In its view, its claims were really directed to novel uses of a natural phenomenon, and should be patent-eligible on that basis.[11]

The district court held the asserted claims were not patent-eligible, and granted summary judgment in Ariosa’s favor.[12] The court reasoned that in addition to fetal DNA that can normally be found in any pregnant woman’s bloodstream, the claims added only “amplifying” DNA from a sample and detecting the presence of paternally-inherited DNA in the sample.[13]  On the record before the court, including admissions from the patent holder’s expert, it was evident that sample preparation and DNA amplification are standard methods well known to those in the art.[14] In the district court’s view, these steps were not “innovative,” but “conventional.”[15]

Sequenom appealed that judgment to the Federal Circuit, arguing that it has not claimed naturally-occurring cffDNA as it exists in the blood stream, but instead, specific uses of compositions of matter that do not exist in nature.[16] First, cffDNA must be separated from its naturally-occurring medium, which is whole blood.  Second, cffDNA is extracellular, and therefore in its natural form, is never “amplified” by normal cellular processes. Third, the amplification process itself generates products that do not exist in nature, because cffDNA is methylated and of various lengths, in contrast to amplification products which are not methylated and have a uniform length. Fourth, the process of detecting the amplification products requires steps that must be performed in a laboratory.[17]

Sequenom argues that those distinctions establish the patentability of its claims, and that because the ‘540 patent establishes a monopoly on specific uses of cffDNA, requiring specific steps, it does not preclude all uses of a naturally occurring composition. According to Sequenom, the ‘540 patent claims should be patent-eligible because they do not prevent all uses of cffDNA, just specific uses involving specific steps.[18]

Sequenom also argues that its claims are distinguishable from those considered by the Supreme Court in Prometheus, because those claims covered methods that physicians had performed before the patent (i.e., administering a therapeutic compound, measuring its metabolites, and adjusting dosages). According to Sequenom, no one had practiced its claims or used maternal serum samples to analyze cffDNA before the patented invention, making the claimed steps an entirely new method.[19] Sequenom urges reversal on the grounds that the district court should have considered whether the entirety of the claim was novel, rather than focusing on the claims’ individual steps.[20]

Sequenom makes the additional argument that an invention may exist when ordinary elements are combined in an extraordinary way.[21]  In Prometheus, the Supreme Court urged that when “claims inform a relevant audience about certain laws of nature,” any additional steps to confer patent eligibility must not simply be steps that are “well understood, routine, [or] conventional” and “already engaged in by the scientific community.”[22] This exhortation is very similar to the Court’s instruction in 2007 in KSR, holding that when a patent “‘simply arranges old elements with each performing the same function it had been known to perform’ and yields no more than one would expect from such an arrangement, the combination is obvious.’”[23]  But the Court has long recognized that patent-eligibility under Section 101 is a separate analysis from patentability for obviousness under Section 103,[24] and litigants must take care to separately address these requirements. 

Pharmacogenomics

Litigants in the Districts of Delaware and Northern California are also embroiled in disputes over the patent-eligibility of nucleic acid-based claims.

In Delaware, Genetic Technologies Limited (GTL) accuses Bristol-Myers Squibb (BMS) of infringing two patents in connection with its pharmacogenomic activities.[25] The patents claim analytical methods that depend on an association of polymorphisms in non-coding genomic sequences with certain measurable traits. GTL accused that BMS infringed those patents by engaging in single nucleotide polymorphism (SNP) studies relating to variability of disease progression or responsiveness to certain drugs.[26]

BMS moved to dismiss GTL’s claims on the grounds that claims of the patents-in-suit are directed to nothing more than a natural correlation involving DNA sequences.[27] BMS argues that there is a natural relationship between genetic polymorphisms in non-coding regions, and those found in coding regions that are also associated with an observable trait in a patient.[28]  According to BMS, any additional steps in the claims are nothing more than “well-understood, routine, conventional activity” previously known in the field.[29]

Even though the claims require man-made primers for amplifying DNA, BMS argues that the claims rely on a considerable amount of information that must already be known. For example, one must know of a gene, know that it is polymorphic, and know that there is a linked polymorphism in non-coding sequence that may serve as a surrogate for a physical trait attributable to the linked, coding DNA.[30]

BMS further argues that DNA amplification cannot be novel because it was well known in the field at the time of the invention.[31] The same holds true of the steps of obtaining a genomic DNA sample, and analyzing or comparing samples. In BMS’ view, the additional steps are either a purely mental act, or a routine step commonly practiced by those in the field, and, in either case, ineligible for patent protection.[32]

GTL, the patent holder, responded to BMS’ motion to dismiss by raising procedural concerns. GTL argues that a dismissal under Rule 12(b)(6) would be improper because the claims must be considered on a claim-by-claim basis, and it has not even identified the claims that will ultimately be asserted.[33] It also relies upon the Federal Circuit’s decision in Ultramercial to argue that a motion to dismiss a patent case on § 101 grounds is “the exception, not the rule,” as a dismissal is only proper when the “only plausible reading of the patent” is that there is clear and convincing evidence of invalidity. Further, such an assessment will ordinarily require claim construction and, further, be “rife with underlying factual issues,” which include whether something was routine or conventional.[34]

As to claim construction, GTL asserts that the claims require a machine, a position at odds with those of the parties challenging the patent.[35] It then addresses the Supreme Court’s decisions in Myriad Genetics, arguing that “man-made DNA” is patent eligible, as are method claims requiring its use.[36] GTL also argues, relying on Myriad Genetics, that an inventor who first discovers the location and sequence of a gene is entitled to claim useful applications involving that sequence.[37]

GTL asserts that analyzing coding and coding sequences according to the claims does not represent a “law of nature” because there is no reliable and immutable correlation between sequences.[38] Moreover, it argues that using a non-coding marker to detect a linked coding marker is a patent-eligible application of the correlation.[39] GTL also argues that the patents’ claims are not ineligible under § 101 because there are other methods that are available to those in the field that are outside of the claims.[40]

Similar to Sequenom’s approach at the Federal Circuit, GTL faults its opponents for attacking the individual claim elements separately, arguing that the combination of the elements in its claims is neither routine nor conventional.[41]

In Northern California, GTL prevailed when faced with a similar challenge on the patent eligibility of its claims.[42]  There, the district court denied Agilent’s motion to dismiss.[43] The court acknowledged that correlations between coding and non-coding regions of genomic DNA are natural laws under Prometheus, but determined that the record was insufficient to determine whether GTL’s claims lacked “sufficient ‘inventive concept.’”[44] The court rejected Agilent’s arguments that the additional claim limitations lacked novelty, because they conflated the analysis of patent eligibility under §101 with patentability under §§102 and 103.[45]

Under Ultramercial, there are several ways in which a claim reciting a natural law may be “meaningfully limited,” so as to be patent eligible under §101:

(1)Does the claim preempt all practical applications of a natural law?

(2)Does the claim contain only insignificant pre- or post-solution activity?

(3)Are the other claim limitations overly-generalized, providing no real direction?

(4)     Does the claim require a particular machine to implement a process, or a particular transformation of matter?[46]

In this case, GTL’s complaint alleged that, at the time its patent was filed, there were numerous, other methods that were available for analyzing genomic variations in coding and non-coding sequences, and its claims were directed to only a subset of methods that require a pair of primers for amplification of those sequences. Accordingly, there was no demonstration that all practical applications of a natural law were preempted.[47]Agilent had not alleged that the amplification step recited in the claims was “insignificant pre-solution activity,” and GTL’s claims, as a whole, were not overly-generalized. Unlike the claims in Prometheus, GTL’s claims require specific steps—amplifying genomic DNA using primers that span a non-coding sequence.[48] Further, GTL had argued that the claims require a “machine,” defined as a “man-made tool,” in the form of primers. Additionally, when practicing the claims, naturally occurring DNA is transformed into man-made, amplified DNA.[49]

In all likelihood, GTL will face additional challenges as to the patent-eligibility of its claims. In denying Agilent’s motion, the court noted that “[w]hether the amplification step in fact meaningfully limits the claimed methods is a question better addressed at a later stage of litigation.”[50]

Looking Forward

After the Prometheus and Myriad Genetics decisions, many practitioners predicted that turbulent times would ensue for patented biotechnological inventions. As courts hear more and more controversies relating to patent eligibility, litigants must confront an evolving legal landscape governing the proper interpretation of 35 U.S.C. § 101.

Litigants are not the only interested parties who must navigate the landscape. On March 4, 2014, the Patent Office issued a Guidance Memorandum that supersedes prior guidance issued after the Supreme Court’s decisions in Prometheus and Myriad Genetics.[51] Under the new guidance, if a claim recites or involves one or more of the judicial exceptions to the four, statutory, patent-eligible subject matter categories, examiners are instructed to consider whether the claims as a whole recites something “significantly different” than the exception.[52]

Patent owners (and those parties that may be accused of infringement) would be wise to seek legal counsel for an up-to-date assessment of how courts view this statute, that, until relatively recently, has largely avoided close scrutiny. It may be quite some time before inventors, those seeking to develop and commercialize inventions, and patent counsel fully comprehend the scope of patent-eligible subject matter in technology areas most relevant to personalized medicine.

Sharon Roberg-Perez and Jamie Kurtz are patent litigators at Robins Kaplan LLP with a focus on biotechnology, medical devices, pharmaceuticals and chemicals.