The Trouble With Diagnostic Method Patents

The last eight years of patent disputes in the courts have set strong but problematic precedents in the area of diagnostic methods in the medical field. Split-court decisions and majority/dissenting opinions have shed new light on what we can expect in the coming decade, both from the various courts, the USPTO, and (hopefully) legislative action.

The Alice/Mayo Test and USPTO Guidance

The two-step test precedent that has ended up determining the outcome of the high-profile patenting cases of the last few years was established in Mayo Collaborative Services v. Prometheus Laboratories, Inc., 566 U.S. 66 (2012), and expanded in 2014 with Alice Corp. v. CLS Bank International, 573 U.S. 208. Essentially, the last decade has been filled with cases of patent ineligibility due to judicial exceptions as they apply to 35 U.S.C §101.

In the Mayo case, Prometheus Laboratories were the exclusive license holders of diagnostic method patents covering a process for determining the speed of metabolizing specific autoimmune treatment drugs in patients and modifying treatment dosage in response to achieve the best outcome. Mayo Collaborative Services had used Prometheus’ diagnostic kits in the past but decided to develop their own independently, upon which Prometheus sued for infringement. The Supreme Court eventually reversed a federal circuit court’s second decision that the claims were patentable, which had itself reversed the district court’s decision to invalidate the patent.

The Supreme Court stated that the relationship between the metabolites and therapeutic effect could not be patented because it qualifies as the judicial exception of “natural law,” and that the subsequent steps “simply tells doctors to engage in well-understood, routine, conventional activity previously engaged in by scientists in the field.” Thus, not enough novelty or groundbreaking application was demonstrated, and this sort of diagnostic process could not be patented.


In Alice, the court applied a similar standard to the “abstract concept” harnessed by software code, ineligible even if it was manifested in a computer. The number of successful “business method” patents fell dramatically after this decision, and the USPTO issued guidance on how to deal with patent eligibility as pertaining to judicial exceptions after the precedent set in the Alice/Mayo decisions, also known as the Alice-Mayo test.

Essentially, this test determines whether the patent as a whole is “directed to” one of the judicial exceptions in step one and if so, the individual claims are analyzed to determine whether they add enough additional and significant elements that can outweigh the judicial exception (step two). In subsequent cases, there has been much debate on whether this test has been applied correctly, and whether this precedent is doing more harm than good.

Athena vs. Mayo

In Athena Diagnostics, Inc. v. Mayo Collaborative Servs., 927 F.3d 1333 (2019), Mayo again questioned the validity of a diagnostic method patent, this time held by Athena Diagnostics, and again won out in the courts. The U.S. Court of Appeals declined an en banc rehearing request last year and seemed to direct a cry for help to the Supreme Court and to lawmakers, citing problems with how §101 has been interpreted to apply to diagnostic methods and the insufficient clarity of the two-step test in this field.

In its majority opinion in the original hearing, the Appeals Court stated that “providing patent protection to novel and non-obvious diagnostic methods would promote the progress of science and useful arts,” which is, after all the idea behind the constitutional aims of the patent system. In their denial of the rehearing, all twelve justices offered varying reasons for why their hands were tied by the Mayo/Prometheus precedent but why they saw the issue as pressing and unsatisfactorily dealt with thus far.

The Supreme Court, somewhat surprisingly to many, declined to grant cert for this case, seemingly leaving the issue to be solved by lawmakers.

The Problem With Diagnostics

There are several tricky aspects to diagnostic method patenting, regarding the business, public health interest, and reading of the Alice-Mayo test in cases such as these.


The medical diagnostic industry spends incredible amounts of money on research and development, which needs to be backed up by some sort of recouping of profit guarantee offered by patents. According to J.D. Houvener, a patent attorney serving Columbia: “make patents harder to obtain, and funding towards this kind of innovation becomes less and less likely. Proponents of revising the law and precedent cite the need for security and reward when new developments are discovered, without which that discovery would see a marked drop.”

Opponents fear that the patent cost and the processes that regular doctors may need to do daily, will slow the implementation of responses to newly discovered diagnostic/disease relationships.

What both camps agree on is the significant power of diagnostic advancements. Diagnostic discoveries help with disease diagnoses, personalized medicine, and aid in stopping illnesses earlier in their progression. Early detection means a more effective and more cost-effective treatment.

However, essentially all diagnostic procedures are based on the relationship between some measurable metabolite or quantitative data in a patient’s blood or measurements and the likelihood of a condition. Technically, all of these relationships are considered laws of nature or natural principles, and thus unpatentable.

The issue with the Alice-Mayo test, in this case, is that the centrality of a newly discovered relationship like this to patent claims lends itself to step one of the processes always being judged to be directed towards that judicial exception. The other claims may not be taken as a whole but individually, and are nearly always existing “standard” techniques of the trade such as immunoprecipitation of metabolites, measuring, and then administering some sort of treatment.

It can be argued that if the metabolite in question is one that has not been used as a diagnostic indicator before, the isolation and measuring are more than just data gathering and represents the novel transformation of materials, but in most cases, this does not get approved. The clever framing of claims could avoid the step one determination and thus the entirety of step two, but the principle of the precedents themselves should probably be improved, and quickly.

Looking Forward

So what comes next? The denial of the Athena cert petition seems like a blow to diagnostic method patent hopefuls, but it may just represent increased pressure on lawmakers to solve the problem on how to patent. In 2018, the Intellectual Property Owners Association (IPO) and the American Intellectual Property Law Association (AIPLA) proposed a solution on how to amend 35 U.S.C. § 101 to solve these issues.

The legislative front of the issue has seen major updates for the first time in nearly a decade, in the form of a draft bill to reform the Patent Act proposed by Senator Coons of Delaware and Senator Tillis of North Carolina. The bill stresses consideration of the entire claim as a whole instead of deconstructing it in terms of eligibility, takes out judicial exceptions, and diminishes consideration of the “well known, routine, or conventional” processes that are involved in most diagnostic methods and yet help make most of them ineligible.

This bill has gone through hearings and is awaiting further revisions, as proponents commend the proposed changes and opponents criticize the reforms as primarily advantaging large companies and potential patent trolls, leading to a monopolization of aspects of the medical industry that should not be patented. Only time will tell how the diagnostic method patenting tangles will be worked out, and how nearly a decade of court precedent may fare in the long term.

Salman Zafar

Originally posted 2020-05-27 07:49:40. Republished by Blog Post Promoter

Your Thoughts

This site uses Akismet to reduce spam. Learn how your comment data is processed.