Sorry you asked? Mayo, Myriad, and the battles over patent-eligibility

Abstract Genetic testing for inherited cancer risk changed dramatically when the US Supreme Court handed down unanimous rulings in Mayo v. Prometheus (2012) and Myriad v. Association for Molecular Pathology (2013). Those decisions struck down claims to methods based on ‘laws of nature’ (Mayo) and DNA molecules corresponding to sequences found in nature (Myriad). Senators Thom Tillis (R-NC) and Christopher Coons (D-DE) introduced legislation that would abrogate those decisions and specify narrow statutory exclusions to patent-eligibility in §101 of the US Patent Act. What would be the consequences of doing so? The Supreme Court decisions coincided with changes in how genetic tests were performed, reimbursed and regulated. Multi-gene sequencing supplanted oligo-gene testing as the cost of sequencing dropped 10,000-fold. Payers dramatically changed reimbursement practices. Food and Drug Administration regulation was proposed and remains in prospect. Databases for clinical interpretation made data freely available, augmenting a knowledge commons. The spectacular implosion of Theranos tempered investment in molecular diagnostics. These factors all complicate explanations of why venture capital funding for molecular diagnostics dropped relative to other sectors. Restoring patent-eligibility would put renewed pressure on other patent doctrines, such as obviousness, enablement and written description, that were not raised in the Supreme Court cases.


I. INTRODUCTION
Four US Supreme Court decisions handed down from 2010 to 2014 changed decadeslong understandings about what can be patented in the USA.The decisions directly affected practices in molecular diagnostics.Two decisions-Mayo v. Prometheus (2012) and Myriad v. Association of Molecular Pathology (2013)-directly addressed diagnostics.They eliminated patent-eligibility for many diagnostic method claims (Mayo) and isolated DNA corresponding to sequences found in nature (Myriad).This created significant uncertainty about what could be patented in molecular diagnostics (and other fields).The decisions disrupted genetic-testing service monopolies on testing one or a few patented genes for various clinical conditions, including inherited risk of breast and ovarian cancer.Before Mayo and Myriad, patented genes had been excluded from many tests for the fear of patent infringement liability.The Supreme Court opened the door to widespread, multigene testing by relieving that threat, so formerly patented genes were included in molecular tests, which became available at a lower cost at many more labs. 1fter the Supreme Court decisions were handed down, several organizations concerned with patent law called upon Congress to abrogate the decisions by revising the definition of patentable subject matter in §101 of the US Patent Act. 2 These efforts were fueled in part by claims that the decisions dampened investment, and therefore innovation, in clinical genomics.Responding to these claims, Senators Thom Tillis (R-NC) and Chris Coons (D-DE) introduced bills in the US Senate to amend §101 of the Patent Act-most recently S. 2140, the Patent-Eligibility Restoration Act of 2023, which would eliminate judicial exceptions to patent-eligibility and replace them with a limited number of statutory exceptions. 3t is unclear whether the bill will become law.At the time of its introduction, GovTrack assessed its odds of enactment at 2 per cent. 4Yet, S. 2140 indicates that interest in a legislative remedy to the uncertainty created by patentable subject matter jurisprudence has not dissipated.It is an appropriate time to examine the practical consequences of statutory replacement of Supreme Court decisions with a more permissive doctrine of patent-eligibility.Here, we do so with a focus on consequences for molecular diagnostics, in particular the implications of the Myriad and Mayo decisions.The paper thus addresses doctrines of patent law but also traces real-world effects on molecular diagnostics.
After providing more detail about the legal background and Congressional response, we turn to the many other factors that changed molecular genetic testing over the same period.We then address some predictable-and some less predictable-consequences if a permissive patent-eligibility statute were to become law.
Our analysis tempers enthusiasm about whether the proposed statutory 'fix' would achieve its intended aims of boosting innovation in molecular diagnostics; indeed, it might well hinder it.We do not address other fields affected by current jurisprudence about patentable subject matter.
Patent-eligibility is indeed more uncertain than before the US Supreme Court weighed in.As explained below, however, patent-eligibility is not the only or necessarily even the most important influence on innovation in molecular diagnostics.Moreover, although the Court decisions affected genetic testing, so did many other events that occurred in parallel.The main effect of the Supreme Court decisions was to enable multi-gene testing to include previously patented genes.That is arguably a major innovation in itself.
Even as the Supreme Court was changing patent jurisprudence, DNA sequencing became much faster and its cost dropped 10,000-fold; public databases to interpret the clinical impact of DNA variants were established, making the information much more broadly available to many more institutions; the way genetic tests were coded, billed, covered, and reimbursed changed dramatically; prospects that FDA would directly regulate genetic tests rose and fell episodically, affecting prospective costs of market entry; and the rise and fall of Theranos cast a pall on molecular diagnostics.
These non-patent events confound understanding the significance of the Supreme Court decisions on genetic testing and predicting whether restoring patent-eligibility would improve investment and innovation in molecular diagnostics.Restoring patenteligibility is unlikely to lead to many, if any, patents on newly discovered human genes in their entirety, however, since there are few left to discover. 5Yet, the proposed statutory language would invite patents on methods and on newly discovered variants of known genes and other DNA elements similar to claims granted in the past.Predicting the consequences of restoring patent-eligibility is further complicated if, as is likely, courts would turn to other doctrines of patent law, such as utility (another feature of §101), novelty ( §102), obviousness ( §103), and enablement and written description ( §112).Litigation strategies would change, but courts might well invalidate patent claims using other patent doctrines to reach the same outcomes.
We do not address the wisdom of shifting to a more permissive patentable subject matter jurisprudence in general, but we question whether reverting to the pre-Mayo patent-eligibility standards will solve the investment incentive problem it is intended to address.In particular cases, for example, startup firms dependent on venture capital (VC), some molecular tests, and methods might benefit from patent incentives.But, systemwide, returning to pre-Mayo patent-eligibility may hinder the improved access to tests and the expanded access and technical innovation that have characterized the past decade of molecular diagnostics.

II.A. Patentable Subject Matter at the US Supreme Court
In the past decade, few events have had a more significant impact on US patent law and practice than the US Supreme Court's four decisions regarding patentable subject matter.
Before these decisions, patentable subject matter was defined largely in the negative: inventions were eligible for patenting unless they were-essentially and when con-4 • Patent-eligibility and molecular diagnostics sidered as a whole-abstract ideas, laws of nature, or natural phenomena. 6Because these exclusions were narrowly defined, inventions were seldom denied a patent based on eligibility, and patents were not often invalidated in court for failure to satisfy the definition of patentable subject matter in §101.
Beginning with Bilski v. Kappos in 2010, however, the US Supreme Court transformed the patentable subject matter doctrine from relaxed gatekeeper to vigilant guard.In Bilski, the Court held that a test previously established by the US Court of Appeals for the Federal Circuit (the appellate court for all US patent decisions) for determining patentable subject matter-the so-called 'machine or transformation' test-provided only an 'investigative clue' as to what constituted patentable subject matter, but did not define it. 7Then, in 2012, the Court introduced a new patenteligibility framework in Mayo that considered whether claims that recite laws of nature 'have additional features that provide practical assurance' that the natural laws are not preempted. 8Applying that framework, the Court held that a method of correlating drug effectiveness to metabolite levels of anti-inflammatory drugs recited an ineligible natural law and did not include features sufficient to make it patent-eligible.
A year later in 2013, the Court held in Myriad that DNA segments whose sequence is found in nature are ineligible products of nature even if they are isolated. 9Myriad attracted considerable public attention, and proved highly controversial within the patent bar. 10 Finally, in 2014 the Court returned to Mayo's eligibility framework in Alice, 11 which the Court described as a 'minor case' that did not break new ground but simply followed Bilski and Mayo.In Alice, the Court elaborated that eligibility depends on whether the patent claim at issue is directed to an ineligible concept, and, if so, whether it adds 'significantly more' that effectively transforms the claim into a patent-eligible application of that concept.Applying this two-step test, the Court held that a method and system for mitigating financial settlement risk were ineligible abstract ideas.
Conventional wisdom about the impact of Mayo and Myriad on genetic testing holds that 'The [Mayo] ruling opened up the field for broader competition and increased access to genetic testing, as it prevented the monopoly on naturally occurring correlations' and '[Myriad] opened up competition and allowed other laboratories to offer testing for BRCA1 and BRCA2 genes, leading to increased accessibility and affordability of genetic testing for hereditary breast and ovarian cancer.The ruling also encouraged further research and development in the field, as scientists and companies were no longer restricted by gene patents.'12When introducing nationally prominent genetic counselor Ellen Matloff a decade out, an interviewer observed 'Few people argue today that the Supreme Court made a mistake.' 13 Yet, it is not so simple.The people who believe the Supreme Court went astray may or may not be few, but their voices are mighty.As noted below, many experts in patent law are dissatisfied with the current state of the law and have urged Congress to assert its authority to define patentable subject matter and restore patent-eligibility.

II.B. Subsequent Case Law
The four Supreme Court decisions upended decades of patent practices in biotechnology.Methods and molecules presumed to be patentable since the 1980s were suddenly in question or clearly invalid.However, the impact on patenting in molecular diagnostics was especially pronounced, given the elimination of an entire category of claims on DNA molecules and related methods.
Myriad Genetics' efforts to enforce claims not challenged in the Myriad case proved unsuccessful in litigation after Myriad was decided. 14Since then, the Federal Circuit has considered a number of diagnostic methods claims under the new patentable subject matter jurisprudence and, to the dismay of some commentators, found them invalid under the Mayo-Alice two-step test.In 2015, for instance, the Federal Circuit ruled that a patent on a non-invasive method for detecting fetal abnormalities from DNA was ineligible despite it being a major advance in noninvasive prenatal genetic testing. 15The Supreme Court's decision not to intervene was called the 'top patent story of 2016,' as the Court demurred despite pleas from the Appeals Court judges to address the uncertainty surrounding patentable subject matter. 16Similarly, in 2018, the same appeals court invalidated a method for detecting Mycobacterium tuberculosis by amplifying specific nucleotides. 17In 2019, the Federal Court held invalid a test to diagnose risk of cardiovascular disease based on levels of a naturally occurring protein. 18Although the Federal Circuit has upheld the patentability of other kinds of method patents-namely, methods of treatment and preparation-'the bottom line for diagnostic patents,' concluded one of its judges, 'is problematic.'19

III.A. Concerns and Backlash
Meanwhile, patent practitioners and legal observers-including David Kappos and Andrei Iancu, two former Directors of the US Patent and Trademark Office (USPTO), as well as prominent members of the patent bar-have turned their attention to Congress.They have weighed in on the impact of the Court's patentable subject matter decisions on patent practice and innovation and described negative consequences for inventors, 20 asserting 'the time has come to amend 35 U.S.C. § 101.' 21 A retired judge of the Federal Circuit described the current state of patentable subject matter jurisprudence as 'intolerable chaos.' 22 Sitting judges concur; the court's denial of en banc review in Athena Diagnostics, Inc. v. Mayo Collaborative Services, LLC included an unprecedented eight separate opinions in general agreement on the 'fraught' state of 'the issue of §101 eligibility, especially as applied to medical diagnostics patents.' 23 We agree there is a problem of uncertainty and inconsistency; we do not, however, find evidence of an innovation deficit for molecular diagnostics that would be addressed by the proposed solution or that expanding patent-eligibility will necessarily improve innovation because of the many other factors that have changed.We explain this position in subsequent sections.
The focus on investment incentives is central because it is the main rationale being proposed for restoring patent-eligibility in the legal scholarship and advocacy reviewed below.Two of the empirical studies point to a reduction in VC funding for molecular diagnostics relative to other fields, and another study finds that small firms have reduced their patent activity in molecular diagnostics.Those studies do suggest a constriction in one pathway to innovation in molecular diagnostics-startup firms securing VC funding based on prospects of patent exclusivity-that was directly affected by the Supreme Court decisions on patentable subject matter.
The empirical studies do not, however, provide evidence of a deficit in overall innovation in molecular diagnostics.Indeed, genetic tests have proliferated over the past decade.Concert Genetics, founded in 2010, tracks clinical genetic testing.Its 2018 'landscape' of genetic testing found over 74,000 tests on the market, which has since increased to 175,000 tests, showing consistent growth in the number of tests year-byyear, and considerable expansion of the market for molecular diagnostics post-Mayo-Myriad. 24In 2017-2018 alone, 801 multi-gene patent tests were introduced.In the wake of the Supreme Court decisions: 'Between January 2014 and March 2018, the total number of panels with 3+ genes including BRCA1/2 grew by more than 10X.' 25 This expansion of testing came primarily through companies that were already engaged in molecular diagnostics in 2013 and past the stage of VC, or through tests that were developed at academic and hospital laboratories that did not depend on VC funding.

III.B. Statutory Proposals
In 2017, the USPTO held a workshop to discuss the new jurisprudence responding to concerns about patent-eligibility.7][28] The recommendations were taken up by Congress in 2019, when Senators Coons (D-DE) and Tillis (R-NC) outlined a proposal to change §101 to 'statutorily abrogate judicially created exceptions to patent eligible subject matter in favor of exclusive statutory categories of ineligible subject matter.' 29 The list of exclusions included fundamental scientific principles, mathematical formulae, economic or commercial principles, mental activities, or 'products that exist solely and exclusively in nature.'The legislative draft proposal was not translated into legislation that year, but now it has been.On June 22, 2023, Sen. Tillis introduced S. 2140 into the 118th Congress, The Patent-Eligibility Restoration Act of 2023, 30 a slight revision of S. 4734 from the 117th Congress. 31S. 2140 proposes to eliminate judicial exceptions to patentable subject matter and replace them with a limited set of statutory exceptions that include 'an unmodified human gene, as that gene exists in the human body.'The Act further clarifies that isolated genes, or genes otherwise altered by human activity, are not 'unmodified' and therefore are eligible for patenting.Another provision focused on abrogating Mayo and Alice, so that 'any process that cannot be practically performed without the use of a machine (including a computer) or manufacture shall be eligible for patent coverage.'For molecular diagnostics, the effect of these provisions would be to overturn the Myriad and Mayo decisions and restore patent-eligibility for both isolated DNA and diagnostic methods.

III.C. Patent Practices Affecting Genetic Testing for Inherited cancer Risk
Before Myriad and Mayo, the typical practice in biotechnology patenting was to include claims on the DNA sequence of a gene that had been isolated and characterized in the laboratory, as well as claims on methods for detecting the variants in its sequence associated with a clinical condition or other biological feature.Method and composition of matter claims often resulted in separate patents on the same underlying discovery or invention.The claims that were most difficult to work around for genetic testing were method claims, 32 most of which were similar to those invalidated by Mayo and Alice.The composition of matter claims invalidated in Myriad were also highly relevant, however, since the process of genetic testing entails making DNA molecules whose sequence mirrors the DNA in a patient's sample. 33

III.D. Sources of Incoherence in Supreme Court Decisions
DNA is both a physical molecule, a chemical, and also the storage and transmission medium for genetic information.DNA's dual nature played a prominent role in the Court decisions in both the USA and a parallel case in Australia.Judge Lourie of the Court of Appeals for the Federal Circuit focused on chemical structure in Myriadhe found the breaking of covalent bonds to create a man-made molecule a structural change sufficient to make isolated DNA molecules patent-eligible. 34 Patent-eligibility and molecular diagnostics • 9 Judge Robert Sweet's district court opinion that invalidated all of Myriad's challenged claims.Judge Sweet's rationale turned on the informational role of DNA, noting that the purpose of a genetic test was to identify the DNA sequence in patients' cells-that is, to extract information.The use of DNA molecules was merely a means to extract that information. 35ustice Thomas's distinction in Myriad, drawing on the arguments presented in amicus curiae briefs filed by the Solicitor General, 36 preserved patent-eligibility for modified DNA-for example, used to make protein therapeutics-while invalidating claims on DNA sequences found in nature.In a separate brief, Eric Lander made that point and added an additional one, directly challenging Judge Lourie's assertion that DNA fragments of genes with broken covalent bonds are not found in nature.He cited data produced by Christina Fan et al., from the Quake laboratory at Stanford, showing that fetal BRCA1/2 fragments were detected in a pregnant woman's blood stream. 37hese influential briefs were cited by six of the Justices during oral argument.
The Supreme Court echoed Judge Sweet's informational argument in invalidating the claims pertinent to screening and diagnosis.Yet the Court also implicitly accepted Judge Lourie's 'DNA as a chemical' logic when it noted that cDNA molecules are structurally (chemically) different DNA molecules, making them patent-eligible. 38The Supreme Court thus created a rule with a bright-line distinction between what can be patented-an engineered DNA molecule-and what cannot-a DNA molecule whose sequence is found in nature.That incoherence has bred confusion: it implies that to become patent-eligible what matters is changing the structure to something not found in nature, and yet the rationale for invalidation of the diagnostic-relevant claims was DNA's information content and not its structure. 39That is, the rationale 35 Judge Sweet's position is close to that adopted in the subsequent case heard by the High Court of Australia in D'Arcy v Myriad, which invalidated the core claim to isolated BRCA1 DNA, and reversed two lower court rulings that had upheld Myriad's claim.36 Brief for the United States as Amicus Curiae in Support of Neither Party, Ass'n for Molecular Pathology v.
Myriad Genetics, Inc., No. 12-398 (2013  for invalidation is based on information content while the justification for patenteligibility of cDNA is a change in molecular structure.Which is it?The Court drew a bright line between patent-eligible and non-eligible molecules, but left no clue for patent-seekers to predict whether information content or chemical structure should determine patent-eligibility.What should a patent prosecutor advise clients seeking patent protection?If the 'found in nature' logic were extended to other molecules, it would eliminate patent-eligibility for valuable therapeutic molecules such as antibiotics and hormones. 40isch and others have argued that novelty, nonobviousness, utility, enablement, and written description could weed out many problematic patent claims, making patentable subject matter dispensable. 41Yet, both Mayo and Myriad rose to the Supreme court as patent-eligibility cases precisely because broad claims survived the prosecution and examination of DNA-based US patents, and litigants chose to focus on patentable subject matter rather than other patent criteria.Perhaps other patent doctrines could be made to do more weeding, but they didn't. The inherently dual nature of DNA as the tangible storage and transmission medium for genetic information will continue to present challenges for patent claim interpretation moving forward.A DNA molecule is created in a laboratory whether through genetic testing to determine the DNA sequence in a person's cells or by using DNA as an intermediary to produce a valuable therapeutic molecule.But those uses are quite different.In genetic testing, any change in the sequence is a mistake that undermines the purpose of testing-to determine the DNA sequence exactly corresponding to that found in the patient's cells.In producing therapeutics, however, DNA is part of a production chain, and almost always entails deliberate modification of the DNA sequence found in nature, which modification would (probably) make it patent-eligible.
The Supreme Court jurisprudence leaves two unresolved conundrums: (1) a twostep framework for assessing patent-eligibility intended to thwart exclusivity over all uses of fundamental discoveries but also, and probably inadvertently, precluding patents on new diagnostic and screening methods that might benefit from patent incentives; and (2) an inscrutable doctrine that sometimes treats DNA as an information storage medium and sometimes as a chemical, with no discernible consistency.

IV. NON-PATENT FACTORS AFFECTING EXPECTED PROFITABILITY OF MOLECULAR DIAGNOSTICS
In addition to the shock of lost patent-eligibility for diagnostic uses of gene patents, several other factors significantly changed practices in genetic testing even as Mayo and Myriad were rising through the courts.The years 2012-2014 saw immense changes in molecular diagnostics, affecting prospects of profitability and incentives for investment and innovation.These factors influence revenues from genetic testing even more directly than patent-eligibility, and thus make it quite difficult to assess the relative importance of patentable subject matter jurisprudence.Coding, billing, coverage, and reimbursement policies for genetic testing would remain in place even if patenteligibility were restored, casting doubt on whether changing the law would remedy any deficit in investment or innovation incentives.The most fundamental technical change during this period was the cost of sequencing, which had profound effects on genetic testing.The famous 'hyper-Moore's curve' figure from the National Human Genome Research Institute documents a powerful technological shock, as Sanger sequencing was replaced by next-generation sequenceby-synthesis methods.The sequence-by-synthesis methods were faster while just as accurate.In the decade 2006-2015, the estimated cost of sequencing a human genome dropped 10,000-fold, from ∼$12,000,000 to ∼$1200 (USD) (see Fig. 1).It is now even lower.While DNA sequencing costs do not include the interpretation or administrative costs of clinical genetic testing, the sequencing cost is a significant component of DNA testing.From its introduction in 1998 until the Supreme Court decision in 2013, Myriad's price (not cost) for comprehensive genetic testing BRCA1/2-its BRACAnalysis ® + BART ® tests-rose from $2400 initially to the range of $4100. 42at is, the cost of generating a whole genome sequence plummeted from five times more expensive than Myriad's two-gene BRCA1/2 test to less than one-third of Myriad's price over the course of a decade (see Fig. 1).Yet the price of the test rose.
Technological progress meant that labs could generate much more genetic information at lower cost.The drop in sequencing cost led to a shift from testing one or a few genes to sequencing many genes (multi-gene testing), all expressed genes (exome sequencing), or the entire genome (whole-genome sequencing).The technological base of genetic testing shifted from Sanger sequencing of DNA segments amplified by polymerase chain reaction to next-generation genome-wide sequencing.But as Judge Bryson observed in his dissent to the Myriad decision at the appeals court level, the genome-wide sequencing strategies infringed claims on individual genes making up the genome. 43That is, multi-gene testing could in theory require accumulating rights on all the genes sequenced: the prospect raised by Heller and Eisenberg's famous 'anticommons.' 44

IV.B.
A Pervasive Shift from Testing One or Two Genes to Multi-Gene, Exome, and Whole-Genome Analysis The shift to widely available multi-gene genetic testing is arguably the most important result of the Supreme Court decisions.Until the Myriad decision, few US labs other than Myriad openly performed clinical testing of BRCA1/2 in the USA for fear of patent infringement liability. 45][48] By 1998, Myriad had 'cleared the market' of US commercial competitors testing BRCA1/2 by sending notification and/or cease-and-desist letters, and through two lawsuits that were settled out of court. 49Myriad's US service monopoly on BRCA1/2 testing evaporated quickly on June 13, 2013, with the Supreme Court decision.Justice Thomas announced the decision just after 10:30 a.m. on June 13, 2013; by noon that day, two firms (Gene by Gene and Ambry) announced they were offering commercial BRCA1/2 testing.Other firms followed.In July, Myriad initiated lawsuits against commercial competitors, invoking claims that Myriad asserted had not been addressed in the Myriad case decided by the Supreme Court.This follow-on litigation did not, however, prevent many other labs from entering the market.By February 2015, those cases had either been dismissed or settled with Myriad signing covenants not to sue for patent infringement. 50he Myriad decision thus cleared the path not only for BRCA1/2 two-gene testing, but more importantly, the inclusion of BRCA1 and BRCA2 in multi-gene panel testing, as well as the reporting of results of BRCA1/2 sequences for exome and whole genome analysis.Before June 2013, BRCA1 and BRCA2 had been left off cancer multi-gene panels offered by commercial competitors. 51fter June 2013, other labs added BRCA1/2 to their panels, offered two-gene testing, or both.As of May 2022, there were 24 companies offering a total of 120 tests for BRCA1/2 as single genes, and 43 companies offering a total of 211 panels for Hereditary Breast/Breast and Gynecological cancers. 52The shift to multi-gene testing was driven by the underlying technology.Instead of stitching together sequences from over 80 'amplicons' (PCR-amplified DNA segments) in the two huge BRCA1/2 genes, labs developed methods for sequencing many genes associated with inherited cancer risk.The methods differed among laboratories, but the underlying shift to multigene testing was pervasive.Even as the Myriad case was progressing, Myriad itself was shifting from its two-gene BRACAnalysis ® (often augmented by its add-on test for large DNA rearrangements, BART ® ), to a 25-gene MyRisk ® test, which debuted in 2013. 53,54This shift to multi-gene testing would not be possible without licensing a multitude of patents, under the gene-by-gene patent strategies that prevailed in the first generation of gene patents.

IV.C. A Shift in how Genetic Tests Were Coded for Coverage and Reimbursement
In addition to a shift to multi-gene testing-generating much more genomic information for the same cost-the drop in sequencing cost also elicited a response from institutions that pay for health care, including genetic tests.The process by which genetic testing labs get paid for their service is intricate and complex, which is described in another paper in this special issue. 55The Secretary's Advisory Committee on Genetics, Health and Society described the payment system in a 2006 report, and recommended a more systematic, evidence-based process for deciding on coverage and reimbursement of genetic tests. 56The relevance here is that the system for reimbursing genetic tests differs starkly from drugs, biologics, other devices, and other medical goods and services associated with strong profitability.Diagnostics have traditionally been a low-margin business.Myriad's blockbuster patent-based model was an outlier, not the norm.
For drugs, biologics, and devices (and when Myriad introduced BRACAnalysis ® ), a manufacturer generally set a price, and then payers responded in a tremendously complicated and opaque process.For most of the period 1998-2012, payment for all genetic tests was based mainly on how many segments of DNA were sequenced.Using the BRCA1/2 example, Myriad billed for analyzing over 80 amplicons-segments of DNA whose sequences were stitched together by computer to assemble the sequence of both genes. 57Tests for other cancer family syndromes, such as Lynch and familial polyposis of the colon, were priced as high or higher on a per-amplicon basis by Myriad and other labs. 58The relatively high prices commanded for genetic tests were in part due to Myriad being among the first tests on the market for genetic testing of a relatively common and high-profile genetic condition, and thus being able to argue for its initial $2400 price (and $700 for the follow-on BART ® introduced in 2006). 59Other labs followed suit.But once in place, reimbursement was formulaic.That is, the Centers for Medicare and Medicaid Services (CMS) paid for lab tests based on the test methodusually based on how many DNA segments were sequenced-and it was the same unit reimbursement rate (per DNA segment) for all labs.Private payers used similar payment criteria. 60his billing and coding framework changed dramatically in January 2013. 61A Clinical Laboratory Fee Schedule was then implemented to pay for genetic tests under Medicare, and other payers used the schedule as a benchmark.The Schedule was based not on method of testing or how many DNA fragments are sequenced, but instead on the clinical condition or the specific gene(s) being tested, with codes assigned to clinical indications.Labs could not set prices for a new product unless it was for an entirely new indication, and they did not control payment levels, because reimbursement was largely determined by the Schedule regularly revised by CMS. 62Private payers followed similar practices.That is, through 2012, testing a cluster of genes entailed adding up the number of DNA amplicons for all the genes in a test and sending the bill to payers; it was gene-agnostic.Starting in 2013, it was based on the specific condition or gene(s) being tested.
In a related change, the coding system that undergirds billing was also in flux.The coding system is operated by the American Medical Association. 63The codes are used by both federal payers (Medicare, Medicaid, Children's Health Insurance Program, Veterans Health Administration, the military health system, the Indian Health Service, and others) and private payers (health plans and insurers).Codes determine billing, which, in turn, determines reimbursement.Whether a test is reimbursed is also subject to a determination that it is covered by that payer's plan.
From 2012 to 2014, payers understood DNA testing was being integrated into clinical care.As costs of sequencing dropped, thousands of scientific papers demonstrated associations between genomic variants and disease risk, and new disease-related genes were discovered.Payers became aware that genetic testing could become pervasive and expensive.And expenditures for genetic testing did increase dramatically. 64hile sequencing costs were plummeting, the costs of encounters with clinicians and interpretation of test results were not; that is, test prices did not drop in parallel to sequencing cost.The prospect of more tests on more people for more conditions loomed as an escalating cost threat to payers.And indeed, by 2021, rising charges to Medicare for genetic testing led the Inspector General to report on the potential for billing scams and unnecessary testing. 65Revising the coding system for genetic tests was intended in part to manage an anticipated tsunami of reimbursement claims.
Even as billing codes changed in 2013, the basis for payment (pricing) also changed.Section 216 of the Protecting Access to Medicare Act (PAMA) of 2014 established a new framework to pay for clinical laboratory tests. 66The CMS (the payer for Medicare, and the federal aspects of Medicaid and Children's Health Insurance Program) would pay based on a periodically revised schedule based on a survey of laboratories.The selection of labs for the pricing survey was contentious and the subject of litigation, not directly relevant here.But from an investor's perspective, this PAMA provision was intended to control lab payments and built on the new coding system that linked tests to clinical conditions.The PAMA statute specified implementation in 2016, but it was delayed until 2018.The rolling PAMA payment adjustment scheme was yet another change, layered on top of the new coding and reimbursement system, again well beyond the control of the service providers-the testing laboratories.
Changes in how tests were coded, billed, covered, and reimbursed directly affected revenue streams.Unlike patented drugs, which can be priced by the company introducing a new drug onto the market, diagnostics became reimbursed under a schedule that gave less pricing control to the lab doing the test.The coverage and reimbursement framework is more similar to generic drugs, treated as commodities to be purchased, a contrast with branded and patented new drugs.
Thus, the basis for revenue flows for molecular diagnostics was disrupted at more or less the same time as the Supreme Court was upending patentable subject matter doctrine.This makes coverage, coding, and reimbursement policies crucial elements of innovation policy, arguably more powerful than patent incentives because they are more directly tied to revenues.Indeed, given that reimbursement schedules and payment levels are much easier to tailor and adjust, they may prove both more powerful and more appropriate ways to create incentives for innovation than patent policy, which strives to be agnostic to particular technologies.The fee schedule for genetic tests is, in particular, arguably the most powerful and specific factor affecting incentives to encourage (or discourage) innovation.

IV.D. Accidental Innovation Policy
While coding, coverage, and reimbursement of genetic tests are powerful drivers of innovation affecting molecular diagnostics, they are generally not framed as innovation policy, but rather as health policy.Conversely, patent policy is generally framed as innovation policy, with little attention to the impact on public health and access.A narrow focus on patents can blind policymakers to other innovation incentives.Several legal scholars have noticed the risk of patent-centric innovation policy.Rachel Sachs reviewed health policies intended to foster access to medical goods and services that nonetheless are powerful influences on innovation, observing health programs are 'accidental innovation policymakers.' 67Lisa Ouellette also pointed to factors beyond patents that are crucial to innovation: research funding, tax incentives, prizes, and data exclusivity; and she cautioned against exclusive focus on patents in innovation policy. 68atents are not the only game in town; indeed for molecular diagnostics, the debate over patent-eligibility may prove a side show compared to coverage and payment policies and regulation: patents may prove important for some specific products, but generally subordinate to the other factors more directly connected to revenue streams: coverage, coding, reimbursement, and regulation.We turn now to regulation.

IV.E. Prospects of FDA Regulation of Laboratory-Developed Diagnostic Tests
Clinical labs in the US are subject to regulation under the Clinical Laboratory Improvements Act of 1988 (CLIA), which is administered by the CMS.This CMS regulatory function is distinct from its role as a payer for Medicare, Medicaid, and the Children's Health Insurance Program (CHIP).CMS certifies labs to ensure they comply with quality measures and have sufficient expertise.CMS does not, however, regulate or approve particular tests.Most DNA-based molecular diagnostics have been classified as 'laboratory-developed' tests, subject only to CLIA oversight by CMS, not FDA approval.They thus occupy an ambiguous regulatory space.As noted, CMS also pays for such tests through Medicare, Medicaid, and CHIP, through its constituent Centers.Thus, CMS is not only the largest single payer, but also the main oversight mechanism for genetic testing.
FDA is not out of the game, however, as prospects of direct FDA regulation of genetic testing have see-sawed for five decades.Test kits that are sold commercially have 67  Patent-eligibility and molecular diagnostics • 17 long been subject to FDA premarket approval under the Medical Device Amendments of 1976. 69But FDA practiced 'enforcement discretion' regarding laboratory developed tests-tests performed in a laboratory as a service but not sold as a product.FDA asserted it had the authority to regulate genetic tests under the device statute, but chose not to do so.This metastable policy was the subject of considerable debate, dating back to the 1980s.N. Anthony Holtzman wrote an appendix to Human Gene Therapy, a 1984 report from the congressional Office of Technology Assessment, that noted the foreseeable rise of genetic testing, and possible need for regulation of genetic tests. 70is analysis expanded into a book, Proceed with Caution. 71The Secretary's Advisory Committee on Genetic Testing, a federal advisory committee, recommended in 2000 that the 'FDA should be the federal agency responsible for the review, approval and labeling of all new genetic tests that have moved beyond the basic research phase.' 72 In 2010, FDA announced an intent to reconsider its enforcement discretion on laboratory-developed tests and hosted a workshop.In 2012, Congress mandated that FDA notify Congress in advance if it intended to regulate laboratory-developed tests.The FDA responded in 2014 with a proposed 'Framework for Regulatory Oversight of Laboratory Developed Tests.' 73 The FDA framework elicited considerable push-back from labs and did not yield regulations. 74A provision mandating pre-market approval of high-risk molecular diagnostics, aligning with the SACGT recommendation from 2000, was briefly included in drafts of what became the 21st Century Cures Act of 2016. 75The diagnostic test regulation provisions were dropped, however, largely because labs, payers, the FDA, and other stakeholders could not reach a consensus.In July 2016, the FDA announced its intent to promulgate regulations, but soon after the 2016 election, withdrew that pledge and instead issued a 'Discussion Paper' in February 2017, early in the Trump Administration. 76he debate continued into the 117th Congress, with bills introduced into both the House and Senate to authorize FDA regulation as the Verifying Accurate Leading-edge IVCT [in vitro clinical test] Development (VALID) Act, 77 summarized by Rachel Sachs in Health Affairs. 78A revised version of the VALID bill was tentatively incorporated into pending FDA legislation, and former FDA Commissioners Scott Gottlieb and Mark McClellan urged Congress to pass it. 79It nonetheless failed to be included in any legislation passed in the final days of the 117th Congress.Senator Rand Paul also introduced a bill in the 117th Congress, the Verified Innovative Testing in American Laboratories (VITAL) Act, stipulating that 'no aspects of laboratory-developed testing procedures shall be regulated under the Federal Food, Drug, and Cosmetic Act (21 U.S.C. 301 et seq.).' 80 The saga continues into the current 118th Congress.
Separately, FDA has the option of pursuing formal regulation under the existing device statute.In June 2023, the Office of Information and Regulatory Affairs in the White House Office of Management and Budget announced the intention to amend FDA regulations pertaining to laboratory-developed tests. 81On October 3, 2023, FDA published the notice of proposed rule-making to regulate such tests, and its FDA's rule was announced on April 29 and published on May 6, 2024. 82An FDA pilot program also began on June 20, 2023, for diagnostics associated with cancer drugs. 83The outcome of FDA regulation is, however, highly uncertain and in any event would take years to roll out. 84The new regulation may well be challenged.Molecular diagnostics will continue to be offered under the shadow of regulatory uncertainty.
FDA regulation directly affects the cost of market entry, and thus investment incentives, profitability, and business models.FDA regulation would likely increase barriers to entry, as it does in drugs.Eisenberg reviewed how FDA's regulatory authority influenced innovation policy in a 2007 review. 85Providing the evidence to enter the market would require expenditure on clinical studies.Yet FDA regulation could also become a source of competitive advantage through non-patent mechanisms, such as data exclusivity-that is, barring competitors from using a firm's data to seek approval of a competing test, thereby enabling protection of R&D investment even without patents.
Hopkins and Hogarth note that FDA regulation might interact with patents on diagnostics similar to features of drug development: patent protection could induce investment in expensive clinical studies of a patent-protected test and preclude free Patent-eligibility and molecular diagnostics • 19 riders from marketing a test without paying for the evidence to prove its clinical utility. 86Scholars at the University of Cambridge have also done a series of studies that point to the value, but also the uncertainty, of patent incentives in molecular diagnostics, summarized in 2022. 87The implication is that making diagnostic patents more accessible again would increase incentives to invest in clinical research, analogous to the introduction of new drugs.Payers often already demand evidence of clinical utility to justify coverage and payment, however (at least for major new tests), so the incremental cost of FDA regulation remains unclear.The need for capital to do clinical studies is a major argument for restoring patent-eligibility, but the interdependency between patent rights and incentives for clinical research remain murky.The policy ping-pong of regulating genetic testing, between the FDA and Congress, may well have affected the expected profitability of molecular diagnostics and thus investor enthusiasm for molecular diagnostics-although it is not clear in precisely how, how much, or even in which direction.
Beyond coverage, payment, and regulation, other factors influenced the practice of molecular diagnostics, changing the dynamics of competition and innovation.The next section addresses informational infrastructure that improved the ability of all laboratories throughout the world to interpret the results of genetic tests.

IV.F. Establishment of Open Data Repositories for Genomic Variants
Genomics as a field evolved a distinctive 'open science' ethos, exemplified by the 1996 Bermuda Principles that called for daily disclosure of DNA sequence data from the labs involved in the International Human Genome Sequencing Consortium (the public Human Genome Project, distinct from the privately funded Celera sequencing effort that emerged in 1998). 88The Bermuda Principles were developed as patents on BRCA1 and BRCA2 were pending.The prospect of patents on individual genes in general, and those two genes in particular, hung like a shadow over the Bermuda proceedings, mentioned frequently by the attendees.Concerns about how patents could impede research were prominent, as detailed by John Sulston in his memoir about the Genome Project, The Common Thread. 89ow patents could affect data flow became apparent during the Myriad litigation.As noted in the discussion of proprietary databases above, Myriad stopped sharing most data with public databases in 2004, five years before the litigation began. 90 ).91 Judge Robert Sweet commented on this perverse effect of the patents in his 2010 district court decision destined for the Supreme Court.Judge Robert Shelby called it out even more forcefully in his 2014 decision about the follow-on lawsuits after Myriad, noting:' . . .[T]he practical result of Myriad's patents has been to hinder or halt follow-up research, data-sharing, patient testing, and the creation of additional and more affordable technologies for BRCA1 and BRCA2 testing . . .Myriad has declined to publicly share critical information regarding its classification of variants, including with its own patients.Instead, Myriad retains that information in a private database.In so doing, Myriad distorts rather than serves the patent system's goal of public disclosure in exchange for exclusive rights.In this way, Myriad has chosen a commercial path that turns much of our patent system policy on its head.'  Trade secrets are often framed as a trade-off with patents-that secrecy would replace public disclosure without the open disclosure of patents.Yet some patents create exclusive rights to collect information, and can become the source of trade secrets that do not expire.Simon and Sichelman note that the adverse effects of such patents could be addressed by research exemptions from infringement liability, independent invention exceptions, patent misuse doctrines, or mechanisms to force disclosure.As a matter of patent policy, as Judge Shelby observed (footnote 91 supra), the trade secret database of cancer-related variants was attributable to the patents.That is, patent exclusivity bred trade secret exclusivity rather than being a trade-off of open disclosure in the patent as a trade-off for temporary exclusive rights.The broad claims granted to Myriad initially covered any way of detecting variants and any DNA molecular of longer than 15 base pairs corresponding to BRCA1 or BRCA2, and this was conventional patent practice at the time; but the patents did not describe the hard work of identifying which variants had clinical significance (beyond those specified in the patent).This is immensely significant, because of the over 70,000 variants in BRCA1/2, only about 7 per cent are understood to confer high risk of cancer.See BRCA Exchange, BRCAe xchange.org.At Myriad, the transient exclusive patent right directly gave rise to permanent exclusive control of data because no other US labs did testing at scale to find new variants.Myriad dramatically improved its ability to interpret genetic test results, but the knowledge was not shared.The proprietary data became a proprietary asset.
Patent-eligibility and molecular diagnostics Even as the Myriad case was being reviewed by the Supreme Court, a movement to bolster open science in genomics and its applications was taking form.The flood of genomic data emerging from cheaper and faster sequencing caused an explosion of scientific and clinical studies.Genomics needed standards and policies to build a 'knowledge commons' to contend with the data deluge. 95Fifty colleagues from eight countries met on January 28, 2013, and agreed to produce a white paper that laid out the arguments for a more systematic framework to foster data-sharing. 96The white paper proposed a framework, which became the Global Alliance for Genomics and Health, now simply GA4GH.97 IV.H.The Global Alliance for Genomics and Health and BRCA Exchange One of the three initial flagship projects of the nascent GA4GH was the BRCA Challenge.As BRCA1/2 testing opened to other labs in June 2013 after Myriad, academic labs and Myriad's commercial competitors had a strong incentive to share data with public databases in order to 'catch up' to Myriad's proprietary database.GA4GH announced the BRCA Challenge to build a data resource for interpreting BRCA variants.Under the Challenge, the BRCA Exchange data resource was established at the University of California Santa Cruz, encouraging data submission from databases and labs around the globe. 98hile the most common variants in BRCA1/2 discovered in the early years of testing could be classified as high or low risk, a very large number of rare variants could not be interpreted because the family and clinical case data were insufficient.In 2015, most of those rare variants could be found in only one of the five major databases in the US and Europe. 99That is, it required searching five different databases-some of which were only available through subscription-to see if a rare BRCA variant had been reported before, and even then, the large tranche of such variants discovered by Myriad were not available.BRCA Exchange enabled a consolidation.It became a 'global resource for variants in BRCA1 and BRCA2.' 100 The first release from BRCA Exchange in October 2016 listed 11,923 variants, 101 fewer than Myriad had in its proprietary database.By March 2024, BRCA Exchange reported 72,467 variants in BRCA1/2. 102ince it is freely available on the Internet, this information is available to any clinical laboratory in the world to assist in interpretation.

• Patent-eligibility and molecular diagnostics
No one lab, health program, or even national health system has sufficient examples of rare variants to interpret them.Data-pooling is an effort to lay the foundation for a collective knowledge commons, a more effective and efficient way to interpret currently uncertain variants than a congeries of data silos.Myriad's trade secrecy ironically created incentives for competitor labs to bolster the public data structures collectively, an unusual instance of a multi-player prisoners' dilemma that resulted in most labs sharing data that could lift all ships. 103he emergence of a data commons countervailing against Myriad's proprietary database meant that Myriad's competitive advantage would dissipate over time.Investors in the firms contributing to a data commons-Myriad's competitors-found data-sharing a collective benefit.They competed on other grounds: speed, accuracy, cost, what genes were tested, how results were interpreted, and clarity of reporting.They could also point to the newly open territory after the Supreme Court decisions as an economic opportunity: a market for genetic testing unencumbered by infringement liability.Firms specializing in genomic analysis such as GeneDx, Invitae, Color, and Ambry responded to that opportunity by doing testing and then sharing data with public databases, as a contrast to Myriad.Large, established laboratory firms, such as Quest and LabCorp (both of which started to do BRCA testing after Myriad) also entered the market and also shared some variant data with public databases.Such datasharing was neither universal nor complete, but it was a dramatic enhancement of the public domain resources for genomic interpretation, and data were not constrained by trade secrecy.Most firms that initiated BRCA testing after Myriad shared data on variants with BRCA Exchange, and with the newly established database for clinical genetics, ClinVar.

IV.I. Establishment of the ClinVar Database for Clinical Genomics
The ClinVar database was established at the National Center for Biotechnology Information in 2013, months after Mayo and before Myriad. 104ClinVar was designed to help clinicians to interpret the clinical significance of genomic variants.ClinVar was not established because of the Supreme Court decisions, but its content was bolstered because of them.
ClinVar did not have data on individual cases, which would raise serious privacy concerns if data could be traced to identifiable individuals, but rather listed variants and the risk classifications assigned to them by the labs contributing the data.ClinVar enabled Patent-eligibility and molecular diagnostics • 23 comparisons between conflicting interpretations, and fostered efforts to improve the reliability and accuracy of variant interpretation. 105,106ClinVar quickly became the go-to resource for genetic testing labs all over the world.
In a global survey of clinical testing labs done in collaboration with GA4GH, ClinVar and gnomAD (a population frequency database at the Broad Institute) were the two databases almost universally consulted when making clinical assessments of genomic variants. 107ClinVar was for all human genes, not just BRCA or cancer-related genes.ClinVar made it progressively more difficult for labs with proprietary databases to say, 'Send your samples to us because only we can interpret them.'The same incentive to build an information commons that gave rise to BRCA Exchange fueled data-sharing with ClinVar, not just for BRCA1/2, but also for clinically relevant genes throughout the genome.

IV.J. The Theranos Effect on Investment in Molecular Diagnostics
Theranos was founded in in California's Silicon Valley in 2003 by Stanford dropout Elizabeth Holmes.Theranos built on the idea that blood from finger-pricks could quickly and accurately detect health conditions at a low cost.Theranos promised a revolution in laboratory diagnostics.Most of its tests were not genetic tests.Instead, Theranos promised a new wave of inexpensive, readily available common clinical tests.It was not a genetic testing company, but its fate affected perceptions of molecular diagnostics, including genetic testing.Theranos's valuation rose from $200 million in 2007 to $9 billion in 2014 and transiently exceeded $10 billion. 108Yet, by 2018, Theranos was worthless and no longer existed.Its spectacular and highly public disintegration is detailed in Bad Blood, by John Carreyrou, the Wall Street Journal reporter who broke the story. 109Theranos's demise began in 2015.Theranos raised relatively little of its money from VC firms despite many efforts to do so, with a few exceptions: Tim Draper (Draper Fisher Jurvetson or DFJ), Donald Lucas (Lucas Venture Group or LVG), and Dixon Doll (DCM). 110Major pharmaceutical and biotechnology firms also generally demurred.Most of Theranos's investors were high-wealth individuals such as Betsy DeVos, Henry Kissinger, Rupert Murdoch, and Larry Ellison.
Until 2015, most news was positive and Theranos's valuation soared.From one perspective, the buzz during Theranos's rise may have piqued interest in medical diagnostics more generally.The countervailing perspective is that investing in a competitor was betting against a well-funded Silicon Valley firm with an A-list of famous Board members and a compelling storyline, led by a photogenic and charismatic CEO building a global persona.That is, Theranos might have discouraged investment in competitors because of its high public profile and expectations.
As the story of deceit and fraud began to emerge and the firm began to fall apart after October 2015, Theranos became a cautionary tale about the difficulty of medical diagnostics and the perils of investing in them.It ultimately led to prison for founder Elizabeth Holmes and her lover and chief operating officer, Sunny Balwani. 111,112For investors and prospective investors, the billions of dollars that evaporated as Theranos sank beneath the waves cast a pall over molecular diagnostics, breeding caution and skepticism.Steve Brozak in Stat News noted 'The Theranos debacle threatens to make investors even less likely to support diagnostic companies in the future, slowing innovation and delaying the introduction of life-saving tests.' 113

V. LEGAL SCHOLARSHIP ON EFFECTS OF PATENT-ELIGIBILITY
Most legal scholarship that attempted to empirically characterize the impact of the four Court decisions on investment in molecular diagnostics did not account for the influence of non-patent factors.David Taylor surveyed 475 VC investors, probing the importance of patent-eligibility in willingness to invest, although at a somewhat less granular level than 'molecular diagnostics' and based on survey data.He concluded patent-eligibility was an important (although not always the most important) factor explaining hesitancy to invest in 'biotechnology, medical device and pharmaceutical industries.' 114 Olson and Ducci note the potential impact reducing patent-eligibility, and propose several options for re-establishing mechanisms to ensure exclusivity in molecular diagnostics. 115n another study, Sasha Hoyt compared VC investment in medical diagnostics compared to other sectors using a differences-in-differences analysis, concluding that 'in the four years following Mayo, [venture capital] investment in disease diagnostic technologies was nearly $9.3 billion dollars lower than it would have been absent Mayo (Fig. 2).' 116,117 Judge Paul Michel, former chief judge of the Court of Appeals for the Federal Circuit, brought a draft of the Hoyt paper to national attention in Stat News 118 arguing that a shortage of Covid tests was due to the Supreme Court's patentable subject matter decisions.His conclusion: 'Blame the Supreme Court,' 119 his remedy: restore patent-eligibility.][122][123] These works are welcome additions to empirical scholarship.But they are all subject to attribution error.They fail to account for the confounds in investment incentives.Instead, they focus exclusively on patents as the only incentives that changed in the multifactorial, storm-torn innovation ecosystem for molecular diagnostics.Patenteligibility was a factor.It was far from the only factor, however, and arguably not the most important.
Another careful empirical study by Rai, Chien and Clark corroborated some of the findings above.In a sophisticated difference-in-differences analysis, they compared molecular diagnostics to a control set of patents, following their trail through the patent office before and after Mayo. 124They found that the criteria for getting patents affecting molecular diagnostics became more stringent post-Mayo.It took longer from application to patent grant, patents were harder to get (more likely not to survive the process), and the main claims were longer.The length of the primary claim was used as a proxy for breadth of claims, with more words and greater specificity yielding narrower claims.They did not find an overall drop in the patent activity for diagnostics, but they did notice less propensity to patent among small firms, corroborating the findings of Hoyt and Taylor's VC hesitancy noted above.They were careful, however, not to claim these findings constituted lower innovation in molecular diagnostics overall, noting the many other factors, such as regulation, coverage, and reimbursement were at play.
The interwoven histories of patent-eligibility and other factors are a cautionary tale about exclusively focusing on patents as innovation incentives.Even the baseline assumption of an innovation deficit is arguably wrong, given the well-documented emergence of thousands of genetic tests and market expansion for genetic testing after Mayo and Myriad. 125wo of the most careful and insightful studies of genetic diagnostics and patents were led by Lori Pressman, who was familiar with technology licensing from her previous experience in the Massachusetts Institute of Technology's licensing office.Pressman first focused on 19 major academic institutions, tracing the rise of US DNA patenting through the 1990s, past its peak in 2001, and pointing to the diversity of Pressman's second study provides indirect evidence about whether patents stimulate innovation in molecular diagnostics.She found a correlation between patent exclusivity and the speed with which a license generated revenue, suggesting that exclusivity, at least in some cases, can accelerate public availability of a diagnostic test.The evidence is not direct, and her conclusion appropriately tentative, but it is an invitation to research that could find more examples and more details about patent exclusivity and molecular diagnostics by paying close attention to licensing terms.That would, however, require access to licensing data, which are generally not shared by either academic institutions or firms.
In our final section, we also point out-based on literature review, interviews with patent scholars and others, 131 as well as discussions we have had with lab directors, policy analysts, and advocates-that policy tools other than patentable subject matter might promote innovation in genetic testing for inherited cancer risk more directly and effectively.Our purpose in this final section is not to generate novel policy proposals or extensively review the literature-each section below is or could be the subject of extensive research-but rather to convey the ideas that emerged from talking to experts about how changes in patentable subject matter jurisprudence might affect molecular diagnostics, especially testing for inherited risk of cancer.

VI.A. A Shift from Injunctive Relief
In the interviews with scholars who had written about patents and molecular diagnostics or those directly engaged in genetic testing, we were surprised to learn that the technological imperative for multi-gene testing was so strong that it might well have emerged despite infringement liability risk, with or without Mayo and Myriad decisions.The medical director at one lab stated flatly, 'We were going to do it anyway,' while acknowledging that it would very likely have entailed a complicated, expensive, and litigious market entry.
It is clear, however, that Mayo and Myriad dramatically reduced the risk of infringement liability, and made the pathway to multi-gene testing faster, cheaper, and less risky.Several patent scholars we interviewed also observed that patent litigation in the context of multi-gene testing, if it reached the courts, might not warrant injunctive relief.Courts might well search for proportionate remedies.That is, patent-holders would not necessarily be able to block competitors, but might instead get reasonable royalties, based on the strength and breadth of their patent claims and the frequency with which variants in the patented genes were tested and reported back to clinicians and patients.Here, patent scholars point to the Supreme Court decision in eBay v. MercExchange, 132 which unsettled the usual, almost automatic remedy of an injunction when patent infringement was found.Proportionate damages instead of injunctions could prevent a patent thicket from blocking the path to multi-gene or whole-genome testing, but would require a complicated-and likely contentious-process for establishing the functional equivalent of a patent pool or multi-party cross-licensing framework ex ante, or for courts to formulae to assess reasonable infringement damages following litigation.

VI.B. Beyond Patentable Subject Matter
In an attempt to describe the overall landscape that might concern those doing multigene testing for inherited cancer risk if patent-eligibility reverted to pre-Mayo status, we searched for gene patents whose claims specifically included at least one of the 27 genes tested by all seven of the US clinical labs contributing the most data to the ClinVar database, plus Myriad Genetics. 133 674 published patent applications.As we reviewed claims on some of the most heavily patented genes, we concluded that a careful infringement risk analysis was a large and complicated task that would nonetheless yield uncertain conclusions.We therefore cannot specify the density of the patent thicket, but the large number of patents and complexity of the claims in those patents did make clear that multi-gene testing could incur infringement liability if patent-eligibility rules reverted to the pre-Mayo-Myriad state.An infringement analysis for a multi-gene, all-exome, or whole genome test would be expensive, laborious, and yet result in considerable uncertainty.

VI.C. Claims that Could re-Emerge
While the Supreme Court decisions invalidated claims infringed by genetic screening and testing for BRCA1/2, not all patent claims were invalidated. 134Claims to full-length complementary DNA were explicitly upheld by the Supreme Court.Most methods of genetic testing did not entail making cDNA, however, so preservation of patenteligibility of modified DNA had little impact on genetic testing.Yet, claims on DNA fragments, primers, probes, and methods that were invalidated by the Supreme Court would be infringed by genetic testing if they became patent-eligible again, and such claims might well resurface if patent-eligibility were restored according to proposed statutory language.Future patents are unlikely to pertain to entire genes because the vast majority of human genes has been disclosed through highly accurate and complete DNA sequencing. 135Broad method claims and claims to isolated DNA variants, however, would again be patent-eligible.Patents granted before Mayo and Myriad exemplify claims that might once again be granted.US Patent 5753441 claimed any method of detecting a variant in germline testing of BRCA1. 136That is, discovering the location and sequence of a gene conferred exclusive patent rights to any way of measuring a variant in that gene for the duration of the patent term for any purpose, including research.These and other method claims were judged invalid by the Court of Appeals for the Federal Circuit under the 2012 had to be reviewed manually because one gene, POLE, flagged unrelated patents (because it is an English word), and to remove other spurious patents.We then started to classify claims on the most commonly tested genes with most patent claims including BRCA1/2 and beyond: MSH6, PMS2, APC, BMPR1A, MUTYH, SMAD4, BARD1, CHEK2, RAD51C, and RAD51D.A spreadsheet with the list of patents, and an explanation of the process are deposited at Open Science Framework.See also Janis Geary et al., Patents Mentioning 27 Hereditary Cancer Genes ( Jul. 26, 2023), https://osf.io/ua9tv.134 Claim 2 in the '282' patent, for example, claimed the complementary DNA for BRCA1.The University of Utah and Myriad characterized the gene and disclosed the cDNA sequence in the patent.Per the Supreme Court decision in Myriad, that claim is not invalid (assuming it meets other tests for validity) and could have been enforced until it expired in 2014.The cDNA claim might have been valuable if someone had developed a therapeutic based on the protein encoded by the BRCA1 gene, analogous to gene patents that underlie therapeutic proteins such as peptide hormones (e.g., insulin and growth hormone) or enzymes.Yet, a full-length cDNA is not generated in genetic testing, so this claim was not relevant to genetic testing.
patentable subject matter, but might simply shift litigation-and uncertainty while case law accumulated-to challenges under other patent law doctrines.That is, restoring patent-eligibility might reduce uncertainty in one domain of patent law while fomenting uncertainty and inviting patent litigation on other grounds without reducing uncertainty overall.Efforts to revise §101 are also apt to elicit strong conflict among constituencies, similar to the inability to find common ground regarding the recommendation for an exemption from infringement liability for second opinions and verification testing.Stakeholders in the fight-firms with starkly different business models and patent preferences, the patent bar, and disease advocacy groups-all have armies of advocates and lobbyists at their disposal.A carefully crafted exclusion from patent-eligibility might be possible to craft that would improve the status quo, but the proposed statutory language does not address the core concerns about patents affecting molecular diagnostics that gave rise to Mayo and Myriad in the first place.Instead, the proposed language in S. 2140 would revert to pre-Mayo-Myriad standards.
Fervent disagreement about optimal policy will likely translate to political conflict in restoring patent-eligibility, and invite future litigation should the law revert to pre-Mayo status.The future thus seems destined to entail continued disagreement, litigation, debate in Congress, and perhaps incremental change on one or more fronts.
In the meantime, the power of molecular diagnostics seems destined to continue to grow, despite financial cross-winds and legal uncertainty, with multi-gene genetic testing, exome testing, and whole-genome analysis becoming more pervasive and integral to management of cancer.The disruption created by Mayo and Myriad may prove propitious in having created a period during which fear of infringement liability receded and pushed the door wide open for multi-gene testing, and genome-wide analysis.We make no predictions about whether current legal constraints on patentable subject matter will continue or be modified by statue-or how statutory change would affect innovation-but clinical practices entailing multi-gene and whole genome analysis will surely continue to advance apace.
human action.It is discerned . . .[a DNA sequence] is a characteristic of the human being from whom the nucleic acid is isolated, a characteristic not shared by all human beings.It has nothing to do with the person who isolates the nucleic acid bearing the mutant sequence' See D'Arcy v. Myriad Genetics Inc, (High Ct.Australia 2015) at 4. 40 Knowles and Prosser note: 'Examples of marketed drugs that have been discovered in nature and then isolated and used in a non-naturally occurring form with important therapeutic uses include penicillin, tetracycline, [erythropoietin], adriamycin, insulin, vincristine, vinblastine, streptomycin, and Vitamin B12'[generic name erythropoietin substituted for trade name Epogen ® that was used in the original].
IV.A. Hyper-Moore-Curve Drops in the Cost, and Increases in the Speed and Accuracy, of DNA Sequencing

Figure 2 .
Figure2.Chronology of events affecting genetic testing for inherited cancer risk.Legend: Timeline of events that influenced VC investments in DNA diagnostics.Supreme Court decisions on patentable subject matter below the line; other notable events above the line.

50
See Jorge Contreras, The Genome Defense: Inside the Epic Battle to Determine Who Owns Your DNA at 324-325 (2021).51 See Cook-Deegan, et al., supra note 46.52 In May 2022 we received a personal communication from Concert Genetics that detailed the total number of genetic tests for hereditary cancer, what category the test was, whether it was a panel or single-gene, what lab offered the test, the name of the test, and what gene targets were included.From these data, we calculated the number of companies offering tests for BRCA1/2.The spreadsheets listing the patents, genes, claims, and other data are on deposit at Open Science Framework at https://osf.io/37x2r/(DOI 10.17605/OSF.IO/37X2R).Which genes were tested by which companies is also detailed in Janis Geary, 53 Myriad's MyRisk ® test was initially 25 genes, then 28, and now (mid-2024) 48 genes associated with cancer of breast, ovary, uterus, colon and rectum, skin, pancreas, stomach, prostate, kidney, lung, endocrine glands, and others.See Hereditary Cancer 48 Gene Panel Associated Syndromes, MyRisk (May 19, 2022), https:// myriad-web.s3.amazonaws.com/myRisk/GeneTable_Syndromes_0422.pdf.54 Judge Shelby's 2014 decision memorandum also recounts part of this chronology.See Memorandum Decision and order denying plaintiffs motion for preliminary injunctions, University of Utah Research Foundation v. Ambry Genetics, Case 2:13-cv-00640-RJS Document 185, US District Court for Utah, 2014.
Since it had86 See Michael M. Hopkins & Stuart Hogarth, Biomarker Patents for Diagnostics: Problem or Solution?30Nat.servicemonopoly on commercial BRCA testing in the US, Myriad got almost all US samples and therefore was in a unique position to discover new variants in the genes associated with breast and ovarian cancer risk.91Myriaddeveloped a state-of-the-art analytical pipeline for interpreting cancer risk of BRCA1/2 variants.92Starting in 2004, those data and methods were protected as Myriad's proprietary assets, under terms of use of their data and web portal.93Cliniciansusing Myriad's online system clicked through an agreement to not share data without permission. 94the open science Breast Cancer Information Core that was established and maintained by Larry Brody and colleagues at the National Human Genome Research Institute.Myriad's public data-sharing changed after Frank's death when Myriad's CEO, Peter Meldrum, became aware of the data-sharing practices.That policy recently changed.Under Myriad's new CEO, Paul J. Diaz, Myriad announced in November 2022 that it would start sharing variant data with the public database ClinVar.See Ellen Matloff, Myriad Genetics (MYGN) To Share Guarded BRCA Variant Information In Public Database, Forbes (Nov.3, 2022), https://www.forbes.com/sites/ellenmatloff/2022/11/03/myriad-genetics-mygn-to-share-guarded-brca-variant-information-in-public-database/?sh=73a55761a772.Myriad piloted a ClinVar submission of over 40 variants in March 2023, and contributed data on 2581 variants in July 2023, the first of regular quarterly deposits.See Personal communication with Karla Bowles, Myriad Genetics (May and August 2023 89 See Georgina Ferry & John Sulston, The Common Thread: a Story of Science, Politics, Ethics, and the Human Genome at 139-147 (2002).90 Myriad published rarely and did not include details sufficient for verification of its conclusions when it did publish.See Robert Cook-Deegan et al., The Next Controversy in Genetic Testing: Clinical Data as Trade Secrets?21 Eur j.Hum.Genet.585 (2013).Conley, et al., also reviewed this history.See Conley, et al., Myriad after Myriad: The Proprietary Data Dilemma, 15(4) N. Carolina J. Law Tech.597 (2014).Myriad's first medical director, Thomas Frank, was an avid data-sharer, and helped establish a See Memorandum Decision, supra note 54.92 See J.M. Eggington et al., A comprehensive laboratory-based program for classification of variants of uncertain significance in hereditary cancer genes, 86 Clin.Genet.229 (2013).93 See Robert Cook-Deegan, et al., The Next Controversy in Genetic Testing: Clinical Data as Trade Secrets?, 21 Eur j.Hum.Genet.585 (2013).94 Simon and Sichelman addressed this anomaly, in which 'Data-Generating Patents' can become the basis for trade secrets.See Brenda M. Simon & Ted Sichelman, Data-Generating Patents, 111 Nw. u. l.Rev. 377 (2017).