Rethinking the Way We Patent Diagnostics

The 2012 Supreme Court decision in Mayo Collaborative Servs. v. Prometheus Labs changed the landscape for patenting diagnostic inventions in the United States. Patent eligibility/ineligibility in the United States stems from two sources: (1) 35 U.S.C. § 101, and (2) judicially-created exclusions, including laws of nature, natural phenomena, abstract ideas, and mental processes. In Mayo, the Supreme Court struck down method of treatment claims for being directed to a law of nature. While Mayo did not create this exclusion, it significantly expanded its applicability to diagnostic-based inventions. This paper explores Federal Circuit case law on patent eligibility of diagnostic-based inventions since the Mayo decision, as well as provides practical guidance on drafting diagnostic patent claims in light of these decisions.

Mayo v. Prometheus: The Test for Patent Eligibility

The claims in Mayo related to optimizing treatment of an immune-mediated gastrointestinal disorder. The method comprised three general steps: (1) administering a drug; (2) determining a particular metabolite level; and (3) adjusting the dose based on that measurement.

The Court set forth a two-part test for determining patent eligibility under the law of nature exclusion. The first step is to determine whether the claim is directed to a law of nature. The scope of the natural law is irrelevant. Even claims directed to natural laws that would narrowly inhibit innovation, such as those in Mayo, are directed to natural laws under this test. If a claim is directed to a law of nature, courts move on to step two. Claims directed to a law of nature may still be patent eligible if there is an “inventive concept” that transforms the claim. In assessing this, courts must look to each claim step as well as the combination of steps. The Federal Circuit previously employed the machine-or-transformation (“MOT”) test. A claim was patent eligible under that test if it (1) was implemented by a machine or (2) transformed an article into a new state. Under Mayo, this test can be “an important and useful clue,” but is not dispositive.

The Court found the claims directed to the “natural” relationship between metabolite concentrations in blood and the likelihood that a particular dose will be effective or harmful. The Court also found the claim lacked the requisite inventive concept. The claim informed medical providers to administer a familiar class of drugs and then to determine metabolite concentrations using conventional techniques. Finally, the claim describes the natural law and instructs medical providers to take it into account. The Court found none of these steps, on their own or in combination, to be inventive.

Patent Claims Found Ineligible Since Mayo

Mayo changed the way courts analyze diagnostic method claims. While the MOT test can be a useful analysis tool, it is no longer dispositive. In the eight years since Mayo, the Federal Circuit has shed some light on how this new framework impacts the patentability of diagnostics.

The Federal Circuit has consistently found claims that utilize conventional methods to exploit natural laws not to be patent eligible. In Ariosa Diagnostics v. Sequenom (2015), the claims related to methods for detecting paternally inherited cell-free fetal DNA (“CffDNA”) in maternal plasma and serum to determine fetal characteristics. The detection method was a polymerase chain reaction (“PCR”) assay. The Court explained that the only new and useful subject matter in the claim was the existence of CffDNA in maternal blood (i.e., a natural law). PCR is a conventional technique that the court found insufficient to transform the claim.

Since Ariosa, the Federal Circuit has heard several cases relating to PCR-based diagnostic methods and invalidated them each time. In Genetic Techs. v. Merial (2016), the Court reviewed claims related to using PCR to detect coding region alleles of a multi-allelic genetic locus. The patent owner argued that the claimed method contained a sufficient inventive concept because it employed PCR in a non-conventional manner. The method involved amplifying a non-coding region sequence in order to detect a linked coding sequence. The Court rejected this argument, describing this as a mental process step that gives the claim a purpose but fails to transform it.

The claims in Roche Molecular Sys. v. CEPHEID (2018) related to a PCR-based method for detecting Mycobacterium tuberculosis (“MTB”). Mirroring the analysis in Ariosa, the Court found this claim was not patent eligible. The claim was directed to the relationship between particular nucleotides and the presence of MTB. While PCR had not previously been used for this particular target, it was nevertheless a conventional technique that failed to transform the claim

The Federal Circuit followed a similar line of analysis in a series of cases between the Cleveland Clinic Foundation and True Health Diagnostics. The claims involved correlating myeloperoxidase (“MPO”) in a patient to a control group to assess the patient’s risk of atherosclerotic cardiovascular disease (“CVD”). Cleveland Clinic I (2017) focused on this comparison and correlation. The Court found the claims directed to the relationship between CVD and heightened MPO levels. The Court invalidated the claim since it merely instructed the use of conventional techniques to exploit this natural law. Cleveland Clinic II (2019) dealt with a continuation patent application from the previous case. The claims there focused on the quantitation of MPO concentration and determining whether the sample’s MPO concentration was elevated compared to a control. The claims did not include the correlation step. Nevertheless, the Court found the claims directed to the same natural law, explaining that elevated MPO concentration compared to a control was just another expression of the natural law. Since the Court found the quantitation methodology conventional, it again found the claims invalid.

Finally, Athena Diagnostics v. Mayo Collaborative Servs dealt with methods for diagnosing neurotransmission or developmental disorders related to muscle-specific tyrosine kinase (“MuSK”). The method involved adding labeled MuSK to bodily fluid, conducting an immunoprecipitation, and looking for the presence of labeled MuSK-antibody complexes indicative of a neurotransmission or developmental disorder. The Court found that the correlation between the presence of MuSK antibodies and certain neurological diseases was a natural law to which the claim was directed. The Court also considered the radioimmunoassay to be a conventional technique and thus found the claims invalid.

Patent Claims Found Patent Eligible Since Mayo

While the above cases highlight the risk for diagnostic method claims post-Mayo, recent cases have shed some light on the types of claims that can survive the new test.

Exergen Corp. v. Kaz USA (2018) dealt with a method of approximating human body temperature using a radiation detector. While the Court found the claim directed to a patent ineligible natural phenomenon, it ultimately held that the claim contained a sufficient inventive concept. In particular, the unconventional body temperature approximation (distinct from skin surface temperature) using a radiation detector was key to the Court’s holding.

Exergen exemplifies a situation where an inventive concept can transform a claim directed to a natural phenomenon. Other recent Federal Circuit cases have demonstrated that claims can involve a natural law without being directed to it. The claims in Vanda Pharm. v. W-Ward Pharm. Int’l (2018) related to a method of treatment for patients with schizophrenia. The claims involved determining whether a patient is a poor CYP2D6 metabolizer via a genotyping assay and administering a particular dose based on the assay’s result. While this claim appears similar to Mayo, there is a key difference. In Mayo, the claims instructed the medical provider to adjust the dose based on a particular metabolite’s concentration, but did not provide a specific dose. In contrast, the claims in Vanda indicated a specific dose range depending on the assay’s result. The Court explained that the inventors recognized a natural relationship, but claimed an application of the relationship rather than the natural law itself.

The Federal Circuit affirmed this principle in Endo Pharm. v. Teva Pharm. USA (2019). This case involved methods for using controlled release oxymorphone for patients with impaired kidney function. The method involved measuring creatinine clearance rate after administering a first dose and administering a new dose based on that measurement. Similar to Vanda, the claims indicated specific dose ranges for the second dose depending on the measured creatinine clearance rate. Therefore, the Court found the claims directed to a new and useful method of treatment rather than an observation or detection.

Illumina v. Ariosa Diagnostics (2020) involved claims for a method of extracting DNA from a pregnant woman and obtaining and analyzing a fetal DNA fragment. The method included three steps: (1) extracting DNA; (2) size-based fractionation to produce fragments of 300 base pairs and less; and (3) fragment analysis. The issue was whether the claims were directed to the natural phenomenon that fetal DNA fragments are smaller than maternal DNA fragments. The Court found that while the claims exploited this natural law, they were directed to the physical process steps for selective fractionation.

Conclusions and Practical Guidance

Patent eligibility of diagnostic claims depends on the two-step test from Mayo. While the MOT test can be helpful, it is no longer controlling. In light of Mayo and subsequent cases, there are several things to keep in mind when seeking to patent diagnostic inventions.

Diagnostic claims are often entwined with natural laws. A diagnosis is generally “results-based” (e.g., drug concentration, antibody concentration, gene expression, metabolite concentration, etc.) and is dependent on bodily function (e.g., metabolism). One effective strategy can be to structure claims as therapeutic claims that apply a diagnostic result. For example, a claim for a method of treating disease state X with therapeutic Y based upon gene expression Z. Importantly, as highlighted in Vanda, this type of claim must include a specific dose range in order to avoid being directed to a law of nature.

Additionally, it may be helpful to claim novel elements explicitly that can qualify as inventive concepts. For example, any novel detection methodology (e.g., machinery, components, or steps) should be included in a claim. If what is being diagnosed is novel (e.g., gene expression profile, protein, antibody, or biomarker), this should also be included in a claim. Claims related to a novel manufacturing process of the diagnostic device itself may also be helpful. In certain cases, “improvement” Jepsen claims may be useful to avoid a claim that monopolizes the natural law. Finally, claiming some features more narrowly in order to clarify the inventive concept may also be necessary. For example, if the invention relies in part on the nature of the sample or the nature of the detection technique it may be important to explicitly highlight that in the claim language.

Overall, while Mayo and the cases since have limited diagnostic claim patentability, all is not lost. Diagnostic methods can still be protected if you keep in mind the general principles from the case the law.

Image Source: Deposit Photos
Author: billiondigital
Image ID: 118711776