Abstract
A
A good example relating to biotechnology is provided by In re Ngai, a 2004 decision by the Court of Appeals of the Federal Circuit (Federal Circuit) involving a patent applicant who attempted to claim a “kit” comprising one or more conventional molecular biology reagents along with instructions for performing a novel method of normalizing and amplifying a population of RNA that would involve the use of the recited reagents. 2 The U.S. Patent and Trademark Office (PTO) allowed some of Ngai's claims that were directed to the method per se, but rejected the corresponding kit claims for lack of novelty, applying the printed matter doctrine and refusing to give the instructions any patentable weight. Ngai argued that the inclusion of useful and nonobvious instructions for using the otherwise-anticipated reagents rendered the kit as a whole patentable, but the Federal Circuit affirmed the PTO's decision to give the instructions no patentable weight under the printed matter doctrine.
In the years since Ngai, the Federal Circuit has overseen a significant expansion of the scope and import of the printed matter doctrine. Historically, the doctrine has been invoked with respect to physical products paired with literally printed material, i.e., printed information on a physical substrate such as paper or a computer-readable medium. Recent judicial decisions have, however, expanded the doctrine to encompass method claims, as well as information that is not embodied in any physical medium, such as advice orally provided by a physician to a patient. In the Federal Circuit's most recent printed matter case, Praxair v. Mallinckrodt, decided in May 2018, the court expanded the doctrine still further, holding that a limitation in a method claim reciting a step of merely thinking about information, i.e., a mental step, constitutes printed matter. 3 The printed matter doctrine is increasingly being applied in the life sciences, particularly with respect to new methods of using known drugs for known indications.
This article begins by briefly outlining the origins of the printed matter doctrine, and then reviews recent developments in the Federal Circuit, focusing on cases involving biotechnology and pharmaceuticals. It then discusses the policy implications of these developments, including the increasingly overlapping relationship between the modern printed matter doctrine and patent eligibility.
The Printed Matter Doctrine
The printed matter doctrine is judge-made law that has historically been invoked by courts and the PTO to prevent the patenting of products whose only point of novelty is printed information bearing no functional relationship to the substrate on which it is printed. 4 In the earliest printed matter cases, the doctrine was applied to claims wherein the alleged point of patentability resides in a novel arrangement or presentation of printed matter. For example, a seminal case decided in 1931, In re Russell, involved claims directed towards a directory in which the names of individuals are indexed in a purportedly novel manner. 5 The would-be inventor argued that “his alleged invention facilitates the finding of names in directories and the like; and that it comprises ‘finished tangible subject matter bearing specifically arranged data or means, combined to produce a novel result.’” The court rejected this argument, finding that the alleged novel arrangement of names was all that distinguished the claim over the prior art, and explaining that the “mere arrangement of printed matter on a sheet or sheets of paper, in book form or otherwise, does not constitute ‘any new and useful art, machine, manufacture, or composition of matter,’ or ‘any new and useful improvements thereof,’ as provided in [Section 101 of the Patent Statute].”
The doctrine subsequently evolved to focus less on the arrangement of printed matter and more on its informational content. The printed matter case law has meandered over time, making it virtually impossible to reconcile all the historical cases into a single coherent doctrine. 6 Nonetheless, by focusing on decisions emanating from the Federal Circuit, it is possible to distill what Professor Kevin Collins has described as the “contemporary core” of the doctrine, which he sees as intended to address:
claims with limitations that recite information with content that is meaningful to the human mind. More specifically, [the doctrine] employ[s] a point-of-novelty analysis to invalidate claims for a lack of novelty under § 102 or nonobviousness under § 103 when the only difference between a claimed invention and the prior art resides in the content of the claimed information. 7
For example, in one of the seminal decisions establishing the contours of the contemporary printed matter doctrine, In re Gulack, the claimed invention was a simple “educational and recreational mathematical device” comprising an endless band (i.e., a loop such as a hatband, ring, or belt) on which is printed an endless series of prime numbers generated by a specific algorithm. The series of numbers generated by the algorithm is “endless” in the sense that if one starts with any number in the series and applies the algorithm to it, it will lead to the next number in the series, and then the next number, eventually resulting in the initial number that one started with, thus beginning the process again and continuing on ad infinitum. The PTO rejected the claim, finding that the endless series of numbers was the sole point of patentability, and that it did not bear a functional relationship to the endless band because the series of numbers could be generated by the algorithm without the use of the band. The Federal Circuit reversed. It concluded that, while the PTO had applied the correct legal test by examining the operative connection between the printed matter (the number series) and the “substrate” (the band), it had erred in not finding a sufficient functional relationship between these two claim elements.
The appealed claims … require a particular sequence of digits to be displayed on the outside surface of a band. These digits are related to the band in two ways: (1) the band supports the digits; and (2) there is an endless sequence of digits—each digit residing in a unique position with respect to every other digit in an endless loop. Thus, the digits exploit the endless nature of the band. 8
Because the printed matter and band shared a functional relationship, the printed matter was given patentable weight in the analysis for nonobviousness. As a point of comparison, the court pointed to a prior art reference cited in the case as an example of a band imprinted with printed matter that did not embody the required functional relationship. The reference suggested imprinting data items, such as historical dates or personages, on a band, but the court found that the only relationship between the data and the band is that the band supports and displays the numbers. The data items are otherwise independent, bearing no direct relationship to the other data entries or to the band itself. Lacking a functional relationship, and in contrast to the claims at issue in Gulack, the printed matter in the prior art reference would not be afforded patentable weight.
Thus, in a nutshell, the contemporary core of the Federal Circuit's printed matter doctrine prescribes a two-part inquiry. The first part asks whether the claim includes a limitation reciting printed matter, i.e., “matter claimed for its communicative content.” 9 If the answer is yes, the second question is whether the printed matter is functionally related to the “substrate” on which it appears. Note that to this day, the Federal Circuit continues to state the functional relationship test in terms of the “substrate” on which the printed matter is applied, even though the doctrine has expanded to encompass information that is orally transmitted or merely mentally processed or considered. Particularly with respect to method claims, the term “substrate” has been interpreted quite broadly, to encompass, for example, a claimed method as a whole.
The Federal Circuit has explained that a claim limitation is directed to printed matter “if it claims the content of information.” 10 In In re DiStefano, for example, the PTO rejected for lack of novelty a claim directed towards a method that enabled an individual to design a web page without requiring that individual to learn hypertext markup language (HTML) or to interact extensively with a web page designer. In assessing the novelty of the claim, the PTO did not afford patentable weight to a “web asset” limitation that the PTO found to constitute printed matter lacking the necessary functional relation to the “substrate.” The Federal Circuit reversed, finding that although the “web assets” recited in the limitation can and likely do communicate some information, the content of the information is not claimed. The board therefore had erred in finding that the origin of the web assets constituted printed matter in the claims at issue, and consequently has erred in assigning their origin no patentable weight in assessing novelty.
Similarly, in In re Lowry, the Federal Circuit held that a computer-based structural database was not printed matter because the data structures “contain[ed] both information used by application programs and information regarding their physical interrelationships within a memory.” 11 Although information directed towards a human audience can constitute printed matter even if it is only “printed” in a computer-readable format, information and instructions directed towards the operation of computer will not be treated as printed matter under the doctrine.
The Federal Circuit recently explained that the reason:
[c]laim limitations directed to the content of information and lacking a requisite functional relationship are not entitled to patentable weight [is] because such information is not patent eligible subject matter under 35 U.S.C. § 101. … While the doctrine's underlying rationale is in subject matter eligibility, its application has been in analyzing other patentability requirements, including novelty under 35 U.S.C. § 102 and nonobviousness under 35 U.S.C. § 103.” 12
Not too long ago, the printed matter doctrine was widely regarded as obscure and largely irrelevant, an arcane non-statutory holdover from an earlier time. Major treatises and patent law casebooks tend to give the doctrine little if any coverage. 13 The Supreme Court has not mentioned it in its recent patent eligibility jurisprudence, even though the Federal Circuit has stated that the printed matter doctrine arises out of Section 101 patent eligibility. In Gulack, the Federal Circuit voiced ambivalence with respect to application of the printed matter doctrine in the context of obviousness, stating that a “printed matter rejection under § 103 stands on questionable legal and logical footing.” 14 Few commentators have afforded it much attention, with the notable exception of Professor Collins, who has written three excellent and insightful articles on the doctrine. 15 In 2004, one commentator went so far as to assert that the printed matter doctrine “rests on shaky legal authority and, in any event, has been whittled away to an archaic common law has-been.” 16 As discussed below, recent Federal Circuit decisions have significantly elevated the doctrine's status, rendering the doctrine anything but a “has-been.”
The Printed Matter Doctrine and Biotechnology
The first decision in which the Federal Circuit applied the printed matter doctrine to a biotechnology patent claim, as far as I am aware, was In re Ngai, decided in 1994 and discussed in the introduction to this article. Ngai's invention was a new method of amplifying and normalizing RNA, and Ngai sought patent claims directed not only toward the method per se, but also product claims directed toward a kit for performing the method, said kit comprising molecular biology reagents to be used in the performance of the method along with instructions as to how to perform the method. The PTO allowed the method claims, but rejected the kit claims under a fairly straightforward application of the contemporary printed matter doctrine: even though the method itself was patentable, the molecular biology reagents were well-known in the prior art for use in other applications, and the instructions for performing the patentable method were deemed printed matter bearing no patentable weight due to an insufficient functional relationship between the instructions and the kit as a whole.
Note that Ngai might be viewed as a slight expansion of the printed matter doctrine. In earlier cases, claims generally explicitly recited some physical substrate on which the printed matter is imprinted, and the printed matter analysis focused on the existence, or lack thereof, of a functional relationship between the printed matter and the physical substrate itself. Ngai's kit claim does not explicitly recite a substrate, and the court, without explanation, simply treats the entire kit itself as the “substrate.” In other words, rather than focusing narrowly on the functional relationship between literal printed matter and the substrate on which it is printed, the court looked for evidence of novelty and nonobviousness in the functional relationship between the information and the claimed product as a whole.
Commenting on the policy underlying this the printed matter doctrine, the court pointed out:
[Were we] to adopt Ngai's position, anyone could continue patenting a product indefinitely provided that they add a new instruction sheet to the product. … Ngai is entitled to patent his invention of a new RNA extraction method, and the claims covering that invention were properly allowed. He is not, however, entitled to patent a known product by simply attaching a set of instructions to that product. 17
I would point out, however, that the consequences of allowing Ngai's kit claim would not have been as dire as the court seems to imply. As a general matter, the kit claim would not constitute a patent on a known product, since it would only cover the known product packaged in combination with the new instructions. On the other hand, in cases where the known product is a drug approved by the Food and Drug Administration (FDA), an analogous claim could function as an extended patent on the product, particularly if potential generic competitors are required by FDA to include the instructions with their product, as discussed in an earlier Holman Report. 18
One might ask, why did Ngai fight so hard for these kit claims, given that the corresponding method claims had been allowed? Ngai (or his attorney) no doubt recognized that the enforcement of method claims can be problematic, particularly when the target of enforcement does not actually practice the method directly, but rather sells a product to be used by customers to perform the method, such as a kit for amplifying and normalizing RNA. Selling the kit would not constitute direct infringement of the method claim, since it is the purchasers of the kit who are actually performing the method, not the supplier of the kit. A company seeking to profit from Ngai's invention by selling a kit for use in the practice of the invention might be found liable for infringement of the method claims under a theory of indirect (contributory and/or induced) infringement, but as a general matter it is easier for a patent owner to prevail if it can establish direct infringement. Ngai's proposed kit claim reciting reagents used in performance of the method along with instructions for performing the method could have addressed this concern by rendering sale of such a kit an act of direct infringement.
In 2010, the Federal Circuit decided two printed matter cases, both involving pharmaceuticals, which marked a more significant expansion of the doctrine. The first of these was King Pharmaceuticals v. Eon Labs, in which one of the asserted patents was based on the discovery that the bioavailability of metaxalone is increased when the drug is administered along with food. 19 Claim 1 of U.S. Patent No. 6,407,128 recited simply the administration of “a therapeutically effective amount of metaxalone in a pharmaceutical composition with food.” There was nothing new about administering metaxalone to patients—the drug had been on the market for many years—and the court found that the claim was inherently anticipated by patients who had taken the drug along with food prior to the purported invention, regardless of whether the prior art recognized that administration with food improved bioavailability.
While the invalidation of Claim 1 did not require application of the printed matter doctrine, the court also upheld a printed matter–based invalidation of Claim 22, which depends from Claim 1 and further limits the method of Claim 1 to situations “wherein the metaxalone is from a container with printed labeling advising that administration with food results in an increase in the [bioavailability] of metaxalone compared to administration without food.” King Pharmaceuticals expanded the recognized scope of the printed matter doctrine by applying it explicitly to a method claim, rejecting the patent owner's argument that the doctrine only applies to product claims.
A more substantial expansion of the printed matter doctrine occurred with respect to Claim 21, which also depends from Claim 1, but unlike Claim 22 did not recite any literal printed matter. Instead, Claim 21 included an additional limitation of “informing the patient that administration of a therapeutically effective amount of metaxalone in a pharmaceutical composition with food results in an increase in [bioavailability] compared to administration without food.” Even though informing a patient is not limited to literally printed information, since it would encompass informing a patient orally, the court found the informing step to be analogous to printed matter, and essentially applied the test for patentability under the printed matter doctrine, explaining:
Although [earlier “printed matter” cases such as Ngai and Gulack] involved the addition of printed matter, such as written instructions, to a known product, we see no principled reason for limiting their reasoning to that specific factual context. Rather, we believe that the rationale underlying these cases extends to the situation presented in this case, wherein an instructional limitation is added to a method, as opposed to a product, known in the art. Thus, the relevant inquiry here is whether the additional instructional limitation of claim 21 has a “new and unobvious functional relationship” with the known method of administering metaxalone with food. 20
The patent owner argued that there is a functional relationship between the “informing” limitation and the method because the limitation increases the likelihood that the patient will take metaxalone with food, thereby increasing the efficiency of the method. However, the court rejected this argument, explaining:
Informing a patient about the benefits of a drug [is not functional because it] in no way transforms the process of taking the drug with food. Irrespective of whether the patient is informed about the benefits, the actual method, taking metaxalone with food, is the same. In other words, the “informing” limitation “in no way depends on the [method], and the [method] does not depend on the [‘informing’ limitation].” In re Ngai, 367 F.3d at 1339 (alterations added). “It is not invention to perceive that the product which others had discovered had qualities they failed to detect.” Gen. Elec. Co. v. Jewel Incandescent Lamp Co., 326 U.S. 242, 249 (1945). 21
In the absence of a functional relationship, the court refused patentable weight to the informing limitation of Claim 21, leaving only the inherently anticipated limitation of independent Claim 1. The court reiterated that the “rationale behind this line of cases is preventing the indefinite patenting of known products by the simple inclusion of novel, yet functionally unrelated limitations.” 22
It bears noting that in the decision below, the district court had invalidated Claim 21 for patent ineligibility under Section 101. This was prior to the Supreme Court's decision in Mayo, and the Federal Circuit chose to affirm on the alternate grounds of anticipation in view of the printed matter doctrine, explicitly leaving for another day the question of patent eligibility of new methods of using a known drug for a known indication. Under Mayo, the patent eligibility of the claim would appear to be an open question, which has been further complicated by the Federal Circuit's decision in Vanda, the subject of a recent Holman Report. 23 In particular, under the Vanda majority's view, Claim 21 could be characterized as directed towards a patent eligible method of using a drug for therapeutic treatment. On the other hand, those siding with the Vanda dissent might plausibly argue that the claim is directed towards a patent ineligible concept, i.e., the improved bioavailability of metaxalone when taken with food, and that the claim does not add anything sufficiently inventive to render it patent eligible. In King Pharmaceuticals at least, invalidation based on the prior art and the printed matter doctrine seems a more straightforward and predictable approach to assessing patentability than the amorphous and indeterminate Mayo framework.
Three months after King Pharmaceuticals, the Federal Circuit published another decision invalidating pharmaceutical claims under the printed matter doctrine, AstraZeneca v. Apotex. 24 The claims at issue were quite analogous to the Ngai claims, i.e., patentable method claims and claims reciting a kit comprising a previously known product combined with instructions to use the product according to the method. As was the case in Ngai, the court upheld the method claims and struck down the kit claims as anticipated by the prior art product, giving the instructions no patentable weight.
The primary difference between AstraZeneca and Ngai is that AstraZeneca's claims are directed to use of an FDA-approved drug, budesonide. The method claims recite the administration of the drug no more than once a day; prior to the invention, it was apparently thought necessary to administer the drug more than once a day. The corresponding kit claims recite a kit comprising two elements: “(1) a budesonide composition or suspension in a sealed container containing 0.05 mg to 15 mg budesonide and a solvent, and (2) a label indicating administration by nebulization in a continuing regimen at a frequency of not more than once per day.”
As noted above, in earlier printed matter cases the meaning of “substrate” was generally clear; it was some physical medium upon which the printed matter was printed, e.g., a ticket, or Gulack's infinite loop. Recall that in Ngai, the kit claim did not specify any particular substrate upon which the instructions were printed, and the court glossed over the fact by simply treating the kit itself as the substrate, arguably loosening the definition of “substrate.”
In AstraZeneca, the patent owner (AstraZeneca) argued that the relevant substrate was the drug, while the generic challenger (Apotex) countered that it was the label explicitly recited in the claim. Apotex would seem to have the better of the argument if the word “substrate” is given its plain literal meaning, which would be consistent with the historic usage of the term pre-Ngai. It would also bolster an argument that the instruction should not be given patentable weight, since it would be difficult to argue that there is a functional relationship between the instruction and the label itself. On the other hand, AstraZeneca's broad and nonliteral interpretation of substrate seems consistent with Ngai, and would also enable AstraZeneca to more plausibly argue the existence of a functional relationship between the information specifying proper use of the drug and the drug itself.
To that end, AstraZeneca argued that, as a general matter, there is a functional relationship between a drug and instructions for the safe use of that drug, particularly when the drug is an FDA-approved product and the instructions are FDA-mandated. The company pointed out that FDA regulations require the label for a drug to include information needed for proper use of the drug, and argued that without the label a physician would be unable to use the drug to safely and effectively treat patients. Given that a drug cannot be approved unless and until FDA approves its label, AstraZeneca argued that the instructions had a functional relationship with the drug composition, rendering drug and a label “inextricably intertwined.”
On appeal, the Federal Circuit affirmed the district court's conclusion that the kit claim is invalid, based on the undisputed fact that the only limitation distinguishing the kit over the prior art was the instruction on the label, and that this instruction was printed matter bearing no functional relationship to the “substrate,” and thus entitled to no patentable weight. The Federal Circuit agreed with the district court that it was immaterial whether the drug or the label was treated as the relevant “substrate,” because in any event the instructions do nothing more than explain how to use the a drug.
The instructions in no way function with the drug to create a new, unobvious product. Removing the instructions from the claimed kit does not change the ability of the drug to treat respiratory diseases. Although AstraZeneca is correct that FDA regulations require a label containing information needed for the safe and effective use of any drug, this is a requirement for FDA approval, not patentability. … Ngai foreclosed the argument that simply adding new instructions to a known product creates the functional relationship necessary to distinguish the product from the prior art. As explained in Ngai, if this court concluded otherwise “anyone could continue patenting a product indefinitely provided that they add a new instruction sheet to the product.” 25
In 2011, the Federal Circuit decided In re Huai-Hung Kao, which involved a claim closely analogous to the claims at issue in King Pharmaceuticals. 26 Kao was an appeal of the PTO's rejection of the following claim directed towards a purportedly novel method of treating patients with extended-release oxymorphone:
A method of providing extended pain relief to patients in need thereof, comprising:
providing information that the average bioavailability of oxymorphone in an oral extended release formulation designed to have a 12[-]hour dosing cycle is increased by at least about 26% for subjects with renal impairment compared to that for healthy subjects, and
providing a therapeutically effective amount of such an extended release oral dosage form of oxymorphone or its pharmaceutically acceptable salt thereof.
It was undisputed that the prior art disclosed a method of providing extended pain relief by administration of a therapeutically effective amount of extended-release oxymorphone. The patent applicant's asserted novel contribution thus resided solely in the first limitation of “providing information” about a previously unknown correlation between renal failure and bioavailability. The applicant argued that the PTO had erred in holding the claim obvious because there was no substantial evidence that the correlation was known in the prior art.
The Federal Circuit affirmed the PTO's rejection of the claim, finding the case “not meaningfully distinct” from King Pharmaceuticals.
Though the correlation between the renal impairment and bioavailability was not known, informing someone of the correlation cannot confer patentability absent a functional relationship between the informing and administering steps. … Just as in King Pharmaceuticals, the informing step does not “transform[ ] the process of taking the drug.” This is because there is no requirement in the claim that the dosage be adjusted in response to the informing step. … Because there is no functional relationship between the two steps in the method, and because the administration of controlled release oxymorphone is squarely present in the prior art, the claim must fail. We agree with our predecessor court that to allow such a claim would effectively remove from the public that which is in the public domain by virtue of its inclusion in, or obviousness from, the prior art. 27
Kao attempted to argue that King Pharmaceuticals should be limited to rejections for anticipation, and that it was improper for the court to extend the printed matter doctrine to a method claim in the context of obviousness. The Federal Circuit easily turned this argument aside, pointing out that in King Pharmaceuticals a method claim was rejected for obviousness under the printed matter doctrine.
Finally, we turn to the Federal Circuit's most recent printed matter decision, Praxair v. Mallinckrodt. 28 The case marks a significant expansion of the doctrine by construing the definition of “printed matter” to include information that is not communicated in either a written or oral form, but that is merely thought about. It also provides an important example of a method claim in which printed matter is found to bear a sufficient functional relationship to the claimed method to be given patentable weight.
The patent at issue in Praxair is directed to the pharmaceutical use of nitric oxide, such as (for example) in methods for improving blood oxygenation in neonates experiencing hypoxic respiratory failure. While the pharmaceutical use of nitric oxide was known in the prior art, the inventors identified a means for identifying patients particularly likely to experience serious adverse effects from treatment with nitric oxide. In other words, the invention relates to a form of personalized medicine. Representative Claim 1 recites:
A method of providing pharmaceutically acceptable nitric oxide gas, the method comprising:
obtaining a cylinder containing compressed nitric oxide gas in the form of a gaseous blend of nitric oxide and nitrogen; supplying the cylinder containing compressed nitric oxide gas to a medical provider responsible for treating neonates who have hypoxic respiratory failure, including some who do not have left ventricular dysfunction; providing to the medical provider: (i) information that a recommended dose of inhaled nitric oxide gas for treatment of neonates with hypoxic respiratory failure is 20 ppm nitric oxide, and (ii) information that, in patients with preexisting left ventricular dysfunction, inhaled nitric oxide may increase pulmonary capillary wedge pressure (PCWP), leading to pulmonary edema, the information of (ii) being sufficient to cause a medical provider considering inhaled nitric oxide treatment for a plurality of neonatal patients who (a) are suffering from a condition for which inhaled nitric oxide is indicated, and (b) have pre-existing left ventricular dysfunction, to elect to avoid treating one or more of the plurality of patients with inhaled nitric oxide in order to avoid putting the one or more patients at risk of pulmonary edema.
29
The purported point of novelty in the claim resides in limitation [3], i.e., the providing of information to a medical provider.
In an inter partes review (IPR) proceeding, the Patent Trial and Appeal Board (PTAB) invalidated this claim under a straightforward application of King Pharmaceuticals and Kao. The “providing information limitation” was treated as printed matter lacking the required functional relationship to the method to be afforded patentable weight, leaving only limitations [1] and [2], which were anticipated by the prior art. Not surprisingly, the Federal Circuit affirmed.
Things get a bit more interesting with respect to dependent Claims 3 and 9. Claim 3 includes a limitation of “evaluating the potential benefit of treating the [neonatal patient] with 20 ppm inhaled nitric oxide vs. the potential risk that inhaled nitric oxide could cause an increase in PCWP leading to pulmonary edema” (the “evaluating” limitation). The PTAB characterized this limitation as a patent ineligible “mental step,” and treated it as printed matter, finding no functional relationship between the evaluating limitation and the claimed method, and thus giving the limitation no patentable weight.
On appeal, the patent owner (Mallinckrodt) claimed that the PTAB erred in failing to give patentable weight to the evaluating limitation of Claim 3, arguing that the question of whether a claim limitation is directed towards a mental step may only be considered in determining patent eligibility, not obviousness. The Federal Circuit rejected this argument, however.
Like the information claimed by printed matter, mental steps or processes are not patent eligible subject matter. And while subject matter eligibility underlies the printed matter doctrine, many of our printed matter cases have arisen in the context of anticipation or obviousness. The printed matter doctrine thus raises an issue where the § 101 patent-eligibility inquiry and the § 102 and § 103 novelty and nonobviousness inquiries overlap. Because claim limitations directed to mental steps may attempt to capture informational content, they may be considered printed matter lacking patentable weight in an obviousness analysis. Accordingly, a limitation that merely claims information by incorporating that information into a mental step will receive patentable weight only if the limitation is functionally related to the substrate.
The evaluating limitation in claim 3 is directed to a mental step that is also printed matter. [It] merely requires a medical provider to think about the information claimed in the providing information limitation of claim 1. … There is no meaningful distinction between claim limitations directed to written information in Kao, Ngai, and AstraZeneca, verbal information in King, and mentally-processed information here. An applicant cannot continue patenting a product indefinitely provided that they add a new instruction sheet, or as we now hold, information together with a purely mental step. 30
Claim 9, one the other hand, which depends from independent Claim 7, provides an interesting example of printed matter that is afforded patentable weight by the Federal Circuit because it bears a sufficiently functional relationship with the claimed method. Independent Claim 7 includes a limitation of “providing to the medical provider … a recommendation that, if pulmonary edema occurs in a patient who has pre-existing left ventricular dysfunction and is treated with inhaled nitric oxide, the treatment with inhaled nitric oxide should be discontinued.” In the assessing the patentability of Claim 7, the PTAB did not give this “recommendation limitation” patentable weight due to the lack of a functional relationship between the recommendation and the claimed method.
However, dependent Claim 9 further recites a limitation of “in accordance with the recommendation of [Claim 7], discontinuing the treatment with inhaled nitric oxide due to the neonatal patient's pulmonary edema.” In other words, Claim 9 recites a limitation that applies the recommendation of Claim 7 to alter the course of treatment. Because the printed matter was actually applied in a non-mental step, the PTAB found it to have a functional relationship with the claimed method, and therefore deserving of patentable weight. The Federal Circuit affirmed this aspect of the PTAB decision.
Of course, a functional relationship between the printed matter and the claimed method does not necessarily render the claim valid; it simply permits the court to give the limitation containing the printed matter patentable weight. The PTAB found Claim 1 to be nonobvious and valid when the recommendation limitation was given patentable weight, but the Federal Circuit reversed on this point, finding that, even though the recommendation limitation bears patentable weight, the claim was still obvious in view of the prior art.
Patent Eligibility and the Printed Matter Doctrine
The relationship between the printed matter doctrine and patent eligibility under 35 U.S.C. § 101 is an interesting one, and it appears to have been modified, or at least clarified, by Praxair. As pointed out by Professor Collins, the printed matter doctrine is a rare example of a doctrine of patentability applied in both the contexts of patent eligibility under Section 101 and novelty and nonobviousness under Sections 102 and 103. 31 In particular, if a patent claim is directed towards printed matter per se, for example, a molecular biology textbook in which the only point of novelty resides in the text of the book, then the claim is patent ineligible under Section 101 through application of the printed matter doctrine. Even though the textbook might appear to literally meet the requirements of patentability set forth in the statute, being a useful and nonobvious article of manufacture, it would be bad policy to permit a patent on subject matter that should be protected, if at all, by copyright. In this context, the printed matter doctrine can be rationalized as a judge-made prohibition against the misuse of patents as “backdoor copyrights.” Professor Collins does not cite to any modern judicial decision applying the printed matter doctrine in the context of Section 101, presumably because the proposition that printed matter per se cannot be patented is so well established that the issue is unlikely to be litigated, but he infers that this remains the PTO's position based on published PTO examination guidelines. 32
As a practical matter, application of the printed matter doctrine is reserved for the context of novelty and nonobviousness. Some recent Federal Circuit decisions do explicitly state that the rationale behind denying printed matter patentable weight in the context of anticipation or obviousness is that printed matter is not patent eligible subject matter under Section 101. Although none of the recent Supreme Court decisions addressing patent eligibility mention the printed matter doctrine per se, they do explicitly identify “mental steps” as a form of patent ineligible abstract idea.
Professor Collins has written a law review article explaining his view that the printed matter and patent eligibility doctrines both place limits on “information that is meaningful to the human mind, … work[ing] together to achieve a single goal: they prevent the discovery of a law of nature from ever being a patentable event in and of itself.” 33 At the time he wrote the article, shortly after Mayo was decided, he characterized the doctrines as complementing one another, with
[t]he printed matter doctrine [restricting] the patentability of the extra-mental information that conveys or disseminates factual knowledge [about the world, and Mayo picking] up where the printed matter doctrine leaves off [, restricting] the patentability of the manifestations of factual knowledge of a law of nature that exist within human minds. If it is simply the mental comprehension of a newly discovered law of nature that distinguishes a claimed process from the prior art, [Mayo] holds that the process is not patentable subject matter. 34
Of course, Professor Collins wrote his article years before Praxair expanded the recognized scope of printed matter to encompass a mental step, and so he assumed the two doctrines were complementary rather than redundant, with the printed matter doctrine applicable to communicated information and Mayo to information that is simply thought about. Post-Praxair, there is significantly more overlap between the two doctrines, since both are applicable to claim limitations reciting a step of thinking (e.g., making inferences) about a newly discovered law of nature. In many cases, particularly in the context of biotechnology and pharmaceuticals, the two doctrines provide alternate vehicles for achieving the same policy objective, and will in many cases yield the same outcome, particularly with respect to diagnostic claims of the type at issue in Mayo.
To better understand the relationship between the post-Praxair printed matter doctrine and patent eligibility, it is useful to consider how the printed matter doctrine might be applied to specific patent claims taken from important patent eligibility decisions. Let us begin with the claim which the Supreme Court considered in Mayo, i.e., Claim 1 of U.S. Patent Number 6,355,623, reciting:
A method of optimizing therapeutic efficacy for treatment of an immune-mediated gastrointestinal disorder, comprising:
administering a drug providing 6–thioguanine to a subject having said immune-mediated gastrointestinal disorder; and determining the level of 6–thioguanine in said subject having said immune-mediated gastrointestinal disorder, wherein the level of 6–thioguanine less than about 230 pmol per 8 x 10
8
red blood cells indicates a need to increase the amount of said drug subsequently administered to said subject and wherein the level of 6–thioguanine greater than about 400 pmol per 8 x 10
8
red blood cells indicates a need to decrease the amount of said drug subsequently administered to said subject.
Using the now well-established Mayo two-part framework, the Court found under Step 1 that the claim was directed towards a natural phenomenon (the correlation between level of drug metabolite [6–thioguanine] and optimal drug dosage), and under Step 2 that the claim did not include enough additional “inventive concept” to be eligible for patent protection. “Inventive concept” is generally lacking when the additional claim elements, alone or in combination, are “well-understood, routine, and conventional to a skilled artisan at the time of the patent.” 35
How would the claim have fared under the contemporary printed matter doctrine? Following Praxair, the mental step of being made aware of the desirability of adjusting drug dosage should be treated as printed matter, and the first question would be whether it bears a new or nonobvious functional relationship with the method as a whole. Based on the Federal Circuit's recent printed matter precedent, it seems highly likely that no functional relationship would be found, because the claim does not recite any limitation applying the information from the mental step, and as a consequence this limitation would not be given patentable weight. The administration and determination steps were apparently not novel and nonobvious at the time the invention, which is presumably why all of the claims were drafted to recite the additional mental step (to distinguish over the prior art). Thus, the mental steps should not be given any patentable weight, leaving only the non-mental steps described in the prior art. The claim appears to be invalid under the printed matter doctrine.
Notice that not only do the two doctrines arrive at the same result, but they do so for essentially the same reason. In both cases, the newly discovered information, i.e., the natural phenomenon, is denied patentable weight, and then the court looks for other limitations in the claim to assess whether they provide the necessary degree of invention to render the claim patentable.
The important difference is that under the printed matter doctrine the court uses well-established principles for assessing the novelty and nonobviousness of the claims in view of the prior art, under the statutory provisions of 35 U.S.C. §§ 102 and 103. In contrast, the inquiry under Mayo entails a search for “inventive concept,” a judge-made standard imprecisely defined by Mayo as excluding that which was “well-understood, routine, and conventional” at the time of the patent. In 2013, Professor Collins predicted that “[o]ver time, [Mayo's] reference to conventional technology may simply come to mean the prior art, but a sui generis definition of conventional technology is also conceivable.” 36 In 2018, however, a panel of the Federal Circuit held in Berkheimer that “well understood, routine, and conventional” does not equate with prior art, but beyond that provided little guidance as to the relationship of the two concepts, stating simply: “[w]hether a particular technology is well-understood, routine, and conventional goes beyond what was simply known in the prior art. The mere fact that something is disclosed in a piece of prior art, for example, does not mean it was well-understood, routine, and conventional.” 37
Significantly, the Berkheimer panel noted quite correctly that “[u]nlike prior art for purposes of §§ 102 and 103, we have no established parameters or guidance for what evidence we can and should consider for inventive concept purposes.” 38
Interestingly, the initial district court decision leading to Mayo found the claims to be patent ineligible because they are directed towards a mental step, i.e., being made aware of the desirability of adjusting drug dosage based on the result of the determination step. When the Federal Circuit initially decided the appeal, it disagreed with this aspect of the district court's decision. Although the appellate court agreed that mental steps are patent ineligible subject matter, and that a claim reciting nothing more than a mental step would be patent ineligible, it went on to point out that all of the Mayo claims recite one or more additional non-mental steps that are patent eligible, such as determining the level of metabolite, and held that the presence of transformative limitations of this type render a claim patent eligible. 39 Of course, the Supreme Court ultimately rejected this rationale, holding that the addition of transformative non-mental steps does not save the claim if those steps are conventional. Presumably, the district court gave the non-mental steps in the claims no weight in the patent eligibility analysis because it assumed that the sole point of patentability in the claim was the mental step, in which case the district court's patent eligibility analysis is essentially the functional equivalent of analysis under the printed matter doctrine.
Now let us turn from Mayo to some claims that have been the subject of patent eligibility decisions in the Federal Circuit. Two of these decisions, Cleveland Clinic and In re BRCA1- & BRCA2, involve claims analogous to the Mayo claim in that they recite a new diagnostic method based on the discovery of a correlation between a biomarker and medically relevant information. Importantly, the claims in these cases do not recite a limitation applying the information generated by the method in a tangible way. In Cleveland Clinic v. True Health Diagnostics, Claim 14 of U.S. Patent Number 7,222,552 is representative of the invalidated claims:
A method of assessing a test subject's risk of developing a complication of atherosclerotic cardiovascular disease comprising:
determining levels of myeloperoxidase (MPO) activity, myeloperoxidase (MPO) mass, or both in a bodily sample of the test subject, said bodily sample being blood, serum, plasma, blood leukocytes selected from the group consisting of neutrophils and monocytes, or any combination thereof;
wherein elevated levels of MPO activity or MPO mass or both in the test subject's bodily sample as compared to levels of MPO activity, MPO mass, or both, respectively in comparable bodily samples obtained from control subjects diagnosed as not having the disease indicates that the test subject is at risk of developing a complication of atherosclerotic cardiovascular disease.
The Federal Circuit found the claim to be analogous to the Mayo claim; in this case, the natural phenomenon is the newly discovered correlation between myeloperoxidase level and the risk of developing a complication of atherosclerotic cardiovascular disease. Since there was nothing new about determining the level of myeloperoxidase in a patient, the court found that the claim failed Step 2 of the Mayo framework by not providing enough additional inventive concept, i.e., determining myeloperoxidase level was found to be routine and conventional in the art.
A similar outcome would have in all likelihood ensued if the court applied the printed matter doctrine to the claim. The court would begin by classifying the “wherein” clause of the patent claim as a mental step and thus a form printed matter. And because the information is not applied in the claimed method, for example, by using the information generated to modify the treatment of the patient, there would be no functional relationship between the printed matter and the method. The wherein clause would receive no patentable weight, leaving the determination of myeloperoxidase level in a test subject as the sole claim limitation for purposes of assessing patentability. Assuming that determination of myeloperoxidase level was anticipated or rendered obvious by the prior art, the claim should be found invalid.
But note that, if it turned out to be the case that determining myeloperoxidase level was novel and nonobvious, the claim could be viable under the printed matter doctrine. This could be the case if, for example, prior to the invention, there was no suggestion in the prior art that would motivate one of skill in the art to determine myeloperoxidase level. Or perhaps determining myeloperoxidase level was technically challenging, and the inventors provided a new and nonobvious way of making determination. Of course, under a scenario where the determining step is novel and nonobvious, the claim should probably also be found patent eligible under Mayo, by virtue of the fact that a novel and nonobvious limitation should generally equate with an inventive concept. Then again, maybe not, for reasons discussed below in the context of cases like In re BRCA1- & BRCA2 and Ariosa.
In an earlier decided genetic diagnostic testing case, In re BRCA1- & BRCA2-Based Hereditary Cancer Test Patent Litig., the Federal Circuit took a different tack to arrive at the same conclusion, i.e., a pure diagnostic testing claim is patent ineligible if it does not recite a limitation applying the information generated by the test. 40 A representative claim recites a method of screening for alterations of the BRCA1 breast cancer gene by amplifying and sequencing a segment of a human subject's BRCA1 gene and comparing it to the corresponding sequence of a wild-type gene. In particular, Claim 8 of U.S. Patent Number 5,753,441 recites:
A method for screening germline of a human subject for an alteration of a BRCA1 gene which comprises comparing germline sequence of a BRCA1 gene or BRCA1 RNA from a tissue sample from said subject or a sequence of BRCA1 cDNA made from mRNA from said sample with germline sequences of wild-type BRCA1 gene, wild-type BRCA1 RNA or wild-type BRCA1 cDNA,
wherein a difference in the sequence of the BRCA1 gene, BRCA1 RNA or BRCA1 cDNA of the subject from wild-type indicates an alteration in the BRCA1 gene in said subject[,]
wherein a germline nucleic acid sequence is compared by amplifying all or part of a BRCA1 gene from said sample using a set of primers to produce amplified nucleic acids and sequencing the amplified nucleic acids.
In In re BRCA1- & BRCA2, an accused infringer argued that the claims were directed to a patent ineligible law of nature, i.e., essentially a comparison of the naturally occurring sequences of mutant and wild-type BRCA1. The Federal Circuit found it unnecessary to decide whether the claim was directed towards a law of nature, because in any event the claim was directed toward an abstract idea, another category of patent ineligible subject matter, and more particularly towards a mental step. Note that in this case and others, the Federal Circuit treats a mental step as a form of abstract idea. The court found that the comparison of wild-type and altered DNA sequences occurs in the human mind and thus constitutes an abstract mental step. The only additional limitation in the claim is the step of amplifying and sequencing all or part of a BRCA1 gene. The court found that amplifying and sequencing DNA was well-known, and thus insufficient to provide the necessary inventive concept under Step 2 of the Mayo framework. It is important to emphasize that the court did not find that the amplification and sequencing of the BRCA1 gene was not inventive, but rather that amplification and sequencing of genomic DNA in general is not inventive, and this is where the outcome might have been different under the printed matter doctrine.
In applying the printed matter doctrine to the claim, the mental step of comparing sequences would be treated as printed matter, and because the information is not applied it does not have the necessary functional relationship to the claim and should not be given patentable weight in an assessment for novelty and nonobviousness. Thus, the question of claim validity would hinge upon whether the remaining limitations—amplification and sequencing of the BRCA gene—were novel and nonobvious. Clearly, amplifying and sequencing genomic DNA was well established in the prior art, and that appears to have been enough to render the claim patent ineligible under the Mayo framework. But under the printed matter doctrine the issue should be whether or not the amplification and sequencing of the BRCA1 gene was nonobvious at the time of invention, not the sequencing of genomic DNA in general. Perhaps the claims would have survived scrutiny under the printed matter doctrine if, for example, the BRCA1 gene was insufficiently disclosed by the prior art, or because its utility as a diagnostic had not been discovered and as a consequence the prior art did not provide sufficient motivation to sequence the gene.
In contrast, in Vanda Pharmaceuticals v. West-Ward Pharmaceuticals International Limited, the Federal Circuit recently upheld the patent eligibility of a diagnostic claim that includes an additional limitation applying the information generated by the test in a tangible way, rather than just mentally being made aware of its implications. 41 Illustrative Claim 1 of U.S. Patent Number 8,586,610 recites:
A method for treating a patient with iloperidone, wherein the patient is suffering from schizophrenia, the method comprising the steps of:
determining whether the patient is a CYP2D6 poor metabolizer by: obtaining or having obtained a biological sample from the patient; and performing or having performed a genotyping assay on the biological sample to determine if the patient has a CYP2D6 poor metabolizer genotype; and if the patient has a CYP2D6 poor metabolizer genotype, then internally administering iloperidone to the patient in an amount of 12 mg/day or less, and if the patient does not have a CYP2D6 poor metabolizer genotype, then internally administering iloperidone to the patient in an amount that is greater than 12 mg/day, up to 24 mg/day, wherein a risk of QTc prolongation for a patient having a CYP2D6 poor metabolizer genotype is lower following the internal administration of 12 mg/day or less than it would be if the iloperidone were administered in an amount of greater than 12 mg/day, up to 24 mg/day.
Vanda was a split decision, with the majority finding that under Step 1 of the Mayo framework the claim was directed towards patent eligible subject matter, i.e., a new and useful method for using a drug. Resolution of the question of patent eligibility at Step 1 obviated the need to look for an inventive concept under Step 2. Writing in dissent, Chief Judge Prost found the claim to be highly analogous to the claims found patent ineligible in Mayo, and that the court should have found under Step 1 of the analysis that the claim is directed towards patent ineligible subject matter, i.e., the discovery and characterization of the correlation between a patient's ability to metabolize the drug, as shown by genetic testing, and the appropriate dosage of the drug. She would have concluded that the claim is patent ineligible, even with the additional step of administering the drug in accordance with the outcome of the genetic test, because in her view this drug administration step is conventional and lacks the degree of inventiveness necessary to satisfy Step 2 of the framework.
I suspect that if the court had applied the printed matter doctrine to Vanda's claim it probably would have been found not invalid. Although the limitation of determining whether the patient is a poor metabolizer could be characterized as printed matter under Praxair, the limitation applying that information in the administration of drug to the patient would seem to provide the necessary functional relationship between the information and the method, thus affording the printed matter patentable weight in an assessment of novelty and nonobviousness. Furthermore, the clearly non-printed matter limitation of administering the drug to a patient in accordance with the outcome of the diagnostic test might be sufficient in and of itself to render the claim patentable.
Now let us turn to some Federal Circuit patent eligibility decisions where the claims are less closely analogous to the Mayo claim, but still relate to biotechnology. The claims at issue in Genetic Technologies and Ariosa are directed toward diagnostic inventions, but are of a different nature than those in the previously discussed cases. Rather than a diagnostic invention based on a newly discovered correlation between a biomarker and medically relevant information, the claims at issue in these cases are directed towards new and improved generic methods of genetic diagnostic testing, not limited to any particular biomarker.
In Genetic Technologies v. Merial, the claimed method was based on the purported discovery that genomic non-coding and coding regions can be genetically linked, allowing for the indirect detection of a specific coding allele of a gene (coding DNA) through the sequencing and analysis of genetically linked non-coding DNA. 42 A representative claim recites:
A method for detection of at least one coding region allele of a multi-allelic genetic locus comprising:
amplifying genomic DNA with a primer pair that spans a non-coding region sequence, said primer pair defining a DNA sequence which is in genetic linkage with said genetic locus and contains a sufficient number of non-coding region sequence nucleotides to produce an amplified DNA sequence characteristic of said allele; and analyzing the amplified DNA sequence to detect the allele.
The Federal Circuit found the claim patent ineligible under the Mayo framework. At Step 1, the court found the claim to be directed towards a law of nature, i.e., the correlation between variations in non-coding regions of the genome and allele presence in a linked coding region. At Step 2, the court found that the balance of the claim did not provide sufficient inventive concept. Amplifying genomic DNA with primer pairs was well-known in the prior art, and thus conventional. The court further found that the analysis step was a mental process step, and since Supreme Court precedent has explicitly identified mental processes as patent ineligible subject matter, the Federal Circuit gave it no weight in its Step 2 analysis for inventive concept.
I would suggest that the claim would also have been struck down if it was analyzed under the printed matter doctrine. The “analyzing” limitation would be treated as printed matter, and since the information generated by the analysis is not utilized in the method it would be treated as nonfunctional and afforded no patentable weight. The amplification limitation was clearly anticipated by the prior art, at least inherently, and under principles of inherency it should not matter whether or not those of skill in the art were at the that time cognizant of the fact that non-coding genomic DNA can be genetically linked to a coding region. 43
In Ariosa Diagnostics v. Sequenom, the claims were directed towards methods of detecting fetal DNA of paternal origin in a maternal blood sample, based on the discovery that non-cellular fetal DNA is present in maternal blood and can be amplified and detected through conventional means. 44 In particular, a representative claim recites:
A method for detecting a paternally inherited nucleic acid of fetal origin performed on a maternal serum or plasma sample from a pregnant female, which method comprises:
amplifying a paternally inherited nucleic acid from the serum or plasma sample; and detecting the presence of a paternally inherited nucleic acid of fetal origin in the sample.
In a controversial decision, the Federal Circuit found the claim to be patent ineligible under Mayo. The court found that the presence of paternally inherited non-cellular fetal DNA in maternal blood is a natural phenomenon, and thus the claimed method “begins and ends with a natural phenomenon.” It then found that the recited steps of amplifying and detecting DNA are routine and conventional, providing insufficient inventive concept to qualify the claim for patent eligibility. In a concurring opinion, Judge Linn found that the method constituted a new and useful application of a natural phenomenon which should be afforded patent protection, but that unfortunately, under his reading of Mayo, the court was compelled to find the claim patent ineligible. Many, including myself, would argue that policy considerations would weigh in favor of finding such an invention patent eligible, and that the Supreme Court likely did not intend its decision in Mayo to deny patent protection to inventions of this type.
It is not clear to me whether a court would find the printed matter doctrine applicable to the Ariosa claim. The detecting limitation would involve a non-mental process for detecting a DNA sequence, such as sequencing or hybridization, but might also be interpreted as including a mental step of being aware of the presence of a paternal DNA sequence. If it were, then that mental step would not appear to be functionally tied to the method, and thus the claim would not be patentable if this awareness was the only aspect of the claim distinguishing it over the prior art. The claim would be patentable, however, if the physical acts of amplifying and detecting this particular form of DNA from maternal blood was novel and nonobvious. This might well have been the case; according to Judge Linn, prior to the invention “no one was amplifying and detecting paternally-inherited cffDNA using the plasma or serum of pregnant mothers.” 45
As was the case in Genetic Technologies, a distinction between the analysis of the Ariosa claims under Mayo versus the printed matter doctrine is that under Mayo the patentability analysis focused on the amplification and detection of DNA in general, whereas under the printed matter doctrine it is the amplification and detection of a specific, newly discovered form of DNA, i.e., non-cellular fetal DNA of paternal origin in a maternal blood sample. To my mind, the latter should be more pertinent to the question of patentability. By basing its decision on the conventional nature of DNA amplification and detection at the time the invention, the court invalidated claims directed towards a highly useful and nonobvious invention of the type the patent system was intended to protect and incentivize. 46
Finally, let us briefly consider CellzDirect, an important Federal Circuit decision that found claims directed towards a biotechnology method patent eligible. 47 The patented method is based on the discovery that hepatocytes that have been frozen and thawed are more stable to subsequent multiple rounds of refreezing and thawing than hepatocytes that have never been frozen and thawed. A representative claim recites:
A method of producing a desired preparation of multi-cryopreserved hepatocytes, said hepatocytes being capable of being frozen and thawed at least two times, and in which greater than 70% of the hepatocytes of said preparation are viable after the final thaw, said method comprising:
subjecting hepatocytes that have been frozen and thawed to density gradient fractionation to separate viable hepatocytes from nonviable hepatocytes, recovering the separated viable hepatocytes, and cryopreserving the recovered viable hepatocytes to thereby form said desired preparation of hepatocytes without requiring a density gradient step after thawing the hepatocytes for the second time, wherein the hepatocytes are not plated between the first and second cryopreservations, and wherein greater than 70% of the hepatocytes of said preparation are viable after the final thaw.
In this case, the Federal Circuit found the claim to be patent eligible, rejecting the argument that the claim is directed towards a natural phenomenon (the ability of hepatocytes to survive multiple freeze-thaw cycles). Instead, it found that the claim is directed towards a new and useful method of preserving hepatocytes. The court's analysis is similar to that which occurred in Vanda, resolving the question of patent eligibility at Step 1 of the Mayo framework by finding the claim directed towards an application of a natural phenomenon rather than the underlying natural phenomenon. Applying the printed matter doctrine, the claim would likewise be found valid, assuming it is novel and nonobvious, since there is no printed matter limitation that could be denied patentable weight.
Policy Implications
As noted above, Professor Collins has posited that the printed matter and patent eligibility doctrines are both aimed at the same fundamental policy concern, i.e., limiting the patenting of “information that is meaningful to the human mind” and “prevent[ing] the discovery of a law of nature from ever being a patentable event in and of itself.” When he wrote his 2013 article making this point, he assumed that the doctrines were complementary and largely nonredundant, based on his understanding at the time that the printed matter doctrine was limited to communicated information, and would not reach noncommunicated information that is merely thought about. Now, post-Praxair, it would seem that the two doctrines overlap significantly, at least with respect to some types of method claims, particularly diagnostic/personalized medicine claims of the type at issue in Mayo. I have shown above that the two doctrines should result in the same outcome with respect to Mayo-type claims, i.e., diagnostic and personalized medicine claims based on the discovery of a new and useful correlation between a biomarker and the medically relevant information. It is worth considering, from a policy perspective, the relative merits of the two different approaches to addressing the policy concerns underlying Mayo.
Critics of Mayo point out that the claims at issue would have in all likelihood been found unpatentable as obvious in view of the prior art, and that the Court's decision to rely instead on an expanded interpretation of the patent eligibility doctrine was unfortunate and unnecessary. The Federal Circuit's recent printed matter jurisprudence could substantially facilitate the use of conventional novelty and nonobviousness analysis to address many of the policy concerns underlying Mayo, potentially in a more appropriate manner.
To a large extent, claim validity analysis under Mayo and the printed matter doctrine mirror one another. In both cases certain subject matter embodied in the claim is identified and bracketed off from the remainder of the claim limitations, either because the subject matter is deemed a patent ineligible concept, i.e., an abstract idea, such as a mental step, or a natural phenomenon/law of nature, or because it is information of the printed matter type. Then the court proceeds to determine whether the remaining limitations of the claim provide sufficient additional inventive concept to render the claim patentable. A major difference is that under the printed matter doctrine, the remaining limitations are analyzed under well-established and statutory-based principles of novelty and nonobviousness, whereas under the Mayo framework, the additional limitations are subject to a requirement of “inventive concept,” a newly minted and vaguely defined judge-made standard that is presumably analogous to nonobviousness, but which could differ in important respects. 48
For example, under Mayo, courts strike down claims for insufficient inventive concept based on a conclusory finding that claim limitations are conventional and well-known, without having to specifically identify and substantiate the basis for this conclusion. Oftentimes, courts have done so without formally construing the claims. 49 In contrast, under the printed matter doctrine a determination that a claim is not patentable is generally based on specific prior art references, and the court is more likely to formally construe claim limitations prior to invalidating the claim, thereby providing a more rigorous, predictable, and procedurally fair approach to assessing claim validity.
Analysis under the printed matter doctrine also tends to focus on more-pertinent prior art than the Mayo framework. In Ariosa, for example, the Federal Circuit found the claims patent ineligible under Mayo because the general techniques of DNA amplification and detection were conventional and well-known, i.e., they existed in the prior art. Under a printed matter–based analysis, the relevant prior art would be narrower and more specific to the claimed subject matter, i.e., amplification and detection of the specific form of DNA recited in the claims. The printed matter approach seems more likely to get to the heart of the relevant issue, which is whether the claim is directed to a new and nonobvious application of newly discovered information about the natural world.
In Mayo, the Court emphasized that while patent ineligible fundamental concepts are unpatentable, specific applications of newly discovered or discerned patent ineligible concepts are, and must remain, patent eligible; indeed, abstract ideas and natural phenomena are generally the raw material for patentable inventions. Mayo also specifically noted that the decision was not intended to preclude the patenting of new methods of using drugs, and presumably, as a general matter, the Court did not intend to dramatically reduce the scope of patentability in the context of biotechnology and medicine. Unfortunately, a strict application of the Mayo framework would unduly threaten the patentability of a host of highly useful biotechnology inventions, as pointed out by the dissent in Ariosa, and thus run counter to good policy and the Court's likely intent. One way to address the concern could be an increased reliance on the printed matter doctrine and a comparison of claims to the prior art, which could, at least in some cases, lead to more appropriate outcomes than a stringent application of the Mayo framework.
The printed matter doctrine would seem to be a better vehicle than Mayo for getting at the question of whether the claim is directed toward a specific application of a patent ineligible concept, as opposed to the concept itself. If none of the claim limitations constitute printed matter, then the claim is likely directed towards a tangible application of a concept or concepts, rather than merely the communication or cognition of information relating to the concept. If the combination of nonprinted matter limitations constitutes a useful, nonobvious, and tangible application of that knowledge, it generally should be patentable, and in all likelihood would be under the printed matter doctrine.
On the other hand, even if the claim does recite a printed matter limitation, the claim will still be valid if the communication or cognition of information is functionally related to the claimed method, and the claim as a whole is useful and nonobvious. As discussed above, in Praxair the court found the necessary functional connection in the fact that the information was actually applied in a tangible way. Thus, the printed matter doctrine effectively sanctions claims which actually apply information, in accord with Mayo's directive that specific applications of fundamental concepts are patent eligible.
In short, public policy might be better served by a move towards more reliance on prior art–based limitations on patentability, taking into account the contemporary printed matter doctrine as appropriate, rather than patent eligibility. The lower courts cannot ignore Mayo and the Supreme Court's patent eligibility jurisprudence, but as I have argued in previous Reports, the open-ended language of Mayo can be interpreted in a relatively permissive matter, as exemplified by the Federal Circuit's decisions in Vanda and CellzDirect. This would allow the courts to rely more heavily on statutory-based requirements of patentability, including novelty and nonobviousness, and the printed matter doctrine could facilitate use of these conventional criteria in a manner that addresses the policy concerns underlying Mayo in a more predictable and appropriate manner, hopefully reducing the likelihood that meritorious patent claims representing a specific application of a natural phenomenon will be inadvertently struck down.