Patent Law & Policy

The Utility Requirement: When Undeclared Policy Choices and Epistemological Barriers Derail the System.

The utility requirement is the hated step-child of patent law. Like Cinderella, at first it seems to have a very marginal role, if one at all, in the game. The big factors in patent are obviousness, claim construction, and the doctrine of equivalents. Of course, novelty (even if not as complex as the others) is crucial as well for all the prior art implications. Utility, on the other hand, seems almost redundant. Scholarship has abundantly explained how the advantages of the utility requirement are limited and, should the requirement not even exist, the market would probably perform the same function anyway. As for its gatekeeper role and its contribution to alleviate the administrative burden on the PTO, some experts suggest that these could easily be accomplished through a more appropriate fee system. So what's left? It is probably time for Cinderella to retire and let the more accomplished dancers lead the show.

Well, maybe not so fast. There is at least one ball where our Cinderella still deserves some attention.

The background music of this dance is the substantial utility requirement, the stage is biotechnology and the significant example addressed by this paper is the Brenner genealogy and, in particular, the In re Fisher decision.

The whole of the utility requirement is contained in the word "useful" within § 101. Through the courts' application the requirement has manifested its function in three different configurations: the operable utility, the beneficial utility and the more relevant one, the practical or substantial utility. The "operable" version simply requires that the invention can actually work according to the inventor's description. The "beneficial" adaptation has its origin in XIX century decisions, which meant to keep at bay inventions considered harmful to society, mainly those related to gambling. It is the "practical" or "substantial" variation that has become more relevant through the interpretation and application of the courts and it is the only kind of utility this paper analyses. The term "substantially useful" in patent law means that the invention needs to have a "concrete, specific, and real-world use" (Nard & Wagner). It is specifically with regard to chemicals compounds and biotechnology industry though that the "practical" utility requirement develops its full potential. This is the stage for our Cinderella. In particular, the application of the requirement becomes more critical in the "upstream" portion of the biomedical and biotech research process (Nard), where the main activity concerns results that are still disconnected from downstream products and applications (those that can ultimately and directly benefit society at large).

The first significant example of this "subject-matter" custom-made application of the substantial utility requirement is the Supreme Court decision in Brenner v. Manson. Brenner articulates an enhanced standard of utility for research intermediates. The Court finds that an "object of scientific research" falls short of the utility threshold because the benefit that the public gains from such accomplishment is too weak and uncertain to justify the quid pro quo inherent in the grant of a patent monopoly. "Unless and until a process is refined and developed to the point where specific benefit exists in a currently available form, there is insufficient justification...a patent is not a hunting license. It is not a reward for the search, but compensation for its successful conclusion". A patent must relate to "commerce rather than philosophy". We can't help but notice that there isn't any trace of a "commerciality" requirement in the statute and that between "commerce" and "philosophy" there is a whole ocean of intermediate conditions. It is not within the functions of the PTO or the Court to establish the probability of business success of the invention. This is the role of the market, which in turn is not an arbiter on patentability.
Almost thirty years after Brenner, the Federal Circuit steps in and remodels the application and the procedural guidelines for the utility requirement in Brana. Besides the shift of the burden of proof (now the PTO has the burden of challenging the veracity of the asserted utility contained in the application), the substantial standard seems to be more lenient towards the applicant: "useful in patent law (...) necessarily includes the expectation of further research and development. The stage at which an invention in this field (pharmaceutical) becomes useful is well before it is ready to be administered to humans." The divergence from Brenner appears undisputable.
About six years after Brana, the PTO enters the game and tries to move the pendulum back towards Brenner. With the 2001 Utility Examination Guidelines, the PTO clearly attempts to raise the bar for the applicant in biotech: the utility has to be "well-established" to avoid rejection. In order to meet the standard of the PTO, the invention must pass a two-prong test: a) a Person Having Ordinary Skills In the Art (PHOSITA) would immediately appreciate why the invention is useful based on its characteristics and b) the utility is specific, substantial, and credible. (§ 2107). Although, because the utility issue is a question of law, Brana's holding seems to be controlling (Nader & Wagner), the oscillation of the pendulum is unequivocal and conveys the tension between the conflicting interests involved in the matter.

In 2005 the Federal Circuit continues the Brenner genealogy with its decision of In re Fisher. Fisher and Lalgudi, who had previously unsuccessfully applied for a patent covering "expressed sequence tags" (ESTs), now challenge the Board of Patent Appeals and Interference's decision in front of the Federal Circuit. Fisher admits that she is not aware of any function of the genes to which EST's corresponds. The Federal Circuit affirms the Board's decision. Chief Judge Michel, writing for the majority, clarifies the meaning of "substantial" and "specific" utility. To meet the qualification of "substantial" utility, the applicant must show a "significant and presently available benefit to the public", while to meet the "specificity" requirement the invention needs to offer "a well-defined and particular benefit". The majority finds that none of the ESTs provides a benefit that can be qualified as substantial or specific. Therefore, the court argues, the claimed ESTs are "mere objects of use-testing" and there is no certainty that any useful discovery can stem out of them. Fisher replies with the analogy of the microscope: her ESTs are research tools that can facilitate scientific progress. The majority doesn't accept this rhetoric argument and emphasizes the difference between the "immediate benefit" of the microscope (magnification) and the absence of a comparable quality in the ESTs for which any use remains a mere unverified hypothesis. The dissent of judge Rader stresses the distinction between "claims for methods" and claims for "research tools" which hold independent utility. Judge Rader embraces the microscope analogy and notices how it is inherent to scientific progress to advance by small incremental steps. More importantly, he comments that the proper gatekeeper for Fisher's claims would be obviousness, an option deeply affected by the Deuel decision which substantially cancelled the hurdle of non-obviousness for DNA sequences and confined the enquiry of the decision-maker to the analysis of the novelty requirement.

In re Fisher raises, among other issues, two general orders of problems that affect the current patent system. One pertains to policy implications associated with a higher or lower utility requirement, the other to the possibility that epistemological factors can determine an industry-specific courts' approach. Let's review them in order.

A relaxed utility requirement is an incentive for early applications whereas an enhanced one favors later applicants (Merges & Duffy). According to the so-called "prospect theory," early patents are considered a stimulus for subsequent additional investments in areas with a high margin of technological advancement. Put differently, early patents operate efficiently when the competition is imperfect and the divergence between potential claimants is more accentuated. A third theory emphasizes how early patents, because they also expire early, increase the share of benefit gained by the public. On the negative side, there is the risk of inducing a "gold rush" effect with associated dissipation of rents especially when the initial pool of participants is aligned on a similar starting line. This scenario, in some scholars' view, is typical of DNA sequences research (Merges & Duffy). Even giving credit to the insight of these theories, it does not appear that early patents and not- enhanced utility standards represent an over-all inefficient choice in the biotech area. The prevalent observation is that early patents give a better opportunity to pools of researchers who can not afford the initial investment that a completed route (from preliminary lab findings to final application) would require and tend to reward strong intuitions and pure talent while offering an opportunity to attract additional funding to "sprint-runner" researchers. The very same homogeneity at the starting line that some theories allege as typical of DNA sequences research, and view as an argument against early patents in biotech, represents, on the contrary, one of the more convincing arguments against a specifically higher standard in the field of DNA research.
The majority in Fisher is not receptive of these arguments and simply endorses the Brenner standard.

The decision in Fisher raises another systemic issue, a fundamental question that patent law has to face and will face even more intensely in the future. Clearly the subject matter of the application can influence the analysis and what would seem clear cut in one field, can appear much more fuzzy in another field. On this ground an epistemological factor seems to vitiate the game. The reality is that patent is a unique kind of law, and a unique kind of discipline indeed. Patent law is a body of rules meant to regulate intellectual accomplishments of the most diverse nature, from the almost intuitive mechanical applications to the most intricate and obscure scientific methodologies. Despite this wide range of contents and specificities, the individuals who legislate, enforce, decide, prosecute, research, teach and study patent law are not experts in many of the underlying subject matters and often do not have proper technical expertise in any of those fields. This (almost inevitable) circumstance induces a spontaneous mechanism. Let's take for example judges: when they consider and decide a case, they have in mind a prototypical model of the idea of invention. Maybe, because more intuitive and familiar, they have in mind a mechanical application or an electronic one. This sort of unconscious blue print influences the way they think about any invention. Biotech is certainly one of the most advanced and specialized ground fields for patent law. It is arguably the field where the knowledge gap between judges and experts is most apparent. Judges seem to surreptitiously level down this inequality applying epistemological criteria that are formed for a much more simplistic world. This process can be unconscious but is never the less highly consequential: the risk being what the Supreme Court fears in Benson, "to freeze process patents to old technologies, leaving no room for the revelations of the new, onrushing technologies." The result in front of us is that courts are playing down the level of knowledge and enhancing the standard of patentability with a subject matter specific approach. They are not allowed to do that, at least not until Congress decides to embrace a subject matter specific approach. The recent words of the Supreme Court in Bilski, quoting Chakrabarty, resonate quite loudly on the point: "courts should not read into the patent laws limitations and conditions which the legislature has not expressed". Judges should instead do the opposite: they should have complete deference for the diverse knowledge and expertise levels typical of different areas and they should refrain from raising the hurdle of utility for specific subject matters. Their choice is not only counterproductive on an efficiency basis; it is also seriously flawed on a pure juridical ground. Introducing industry specific standards, as for general category bans, is too relevant of a policy determination to be entertained by the judiciary on a sole journey. These are decisions that are going to affect every-day life for all of us, therefore they need a wider consensus that only the legislative process can express.

It would be unfeasible to have different patentability rules for different subject matters (also the costs of such a system would raise to prohibitive levels), but it is apparent that some issues, like the utility requirement can be perceived in a different light depending on the subject to which the invention pertains. At this point a question remains to be addressed: if there is agreement on the necessity of uniform patentability rules that are not subject-specific, and if under a common standard of utility, we either risk to allow inventions that are not particularly finalized or to exclude inventions that (as in the biological area) could represent a significant progress in the state of the art, although not one with immediate meaningful applications, which option do we want to endorse? Because both outcomes are suboptimal, the decisive criterion becomes what drawbacks we want to consider more inefficient from a policy perspective. In my opinion the less inefficient alternative is to depart from a higher utility standard for biotech and allow the benefit of patent exclusivity to inventions that can lack an immediate applicative translation. The risk of discouraging researchers who have the talent but not the necessary infrastructures to pursue their inventions and who must wait until a possible break through discovery gets to the stage of practical and significant applications tips the balance in favor of a standard utility requirement. It is instead more appropriate to allow the courts to decide if a particular scientific endeavor represents a significant enough leap from the status of the prior art without "industry-specific" interpretative devices. This can be properly done with a non-enhanced utility requirement combined with an orthodox use of the obviousness requirement. From this perspective, a refutation of the unfortunate Deuel holding should be among the Court of Appeals priorities. Besides an overdue return to a standard use of the obviousness requirement, judges have at their hand another device which is perfectly apt to address the different knowledge levels involved in fields with such high levels of specialization as biotech: the interpretative device that shapes the majority of the critical evaluations in patent law: the PHOSITA. This is a flexible instrument that, as a professional standard of competence, is supposed to reflect the real expertise in the field (Burk & Lemley). The PHOSITA is indeed inherently subject-matter specific. His approximation to the real field of the subject matter of the invention is an incredibly powerful interpretative device and plays a key role in correctly assessing the patentability requirements. In biotechnology, a more correctly construed PHOSITA would see utility more often than a currently utilized one does, and would not see non-obviousness and novelty as often as the current one does.