Patent protection for CRISPR: an ELSI review

Abstract

Much has been written about the power of CRISPR—the workhorse genetic-editing system first elucidated in 2012—and the public's interest in it, both as a piece of science and an ethical battleground.1 But there has also been extensive interest in the variety of intellectual property issues surrounding CRISPR, including a heated patent dispute between two of the technology's originators, Jennifer Doudna (UC Berkeley) and Emmanuelle Charpentier (Max-Planck), on one side, and Feng Zhang (Broad Institute) on the other.2 While the intellectual property disputes concerning CRISPR are far from over—indeed, like Tolstoy's War and Peace, new characters central to the dispute continue to materialize3—five years of hindsight has given some perspective on their ethical, legal, and social implications. This brief essay reviews several of these issues: (i) difficulties with interinstitutional collaborations, especially for lucrative ‘translational’ technologies; (ii) the rise of for-profit ‘surrogate’ companies to manage university licensing; and (iii) the use of patents as a means of private governance to prevent potential abuses, such as ‘gene drives’, seed-saving restrictions for agriculture, and germ-line human engineering. It concludes with several observations—and prescriptive recommendations—for patent protection related to academic, collaborative, cutting-edge research. Since the first U.S. patent applications were filed for an engineerable CRISPR system in 2012, the IP landscape has become significantly more crowded, with several researchers controlling a few significant battlements. To date, three groups of scientists have emerged as holders of foundational patents covering CRISPR-Cas9: Jennifer Doudna (UC Berkeley) and Emmanuelle Charpentier (now at Max-Planck, but at Umeå University, Sweden, at the time of her contribution to the invention), Feng Zhang (Broad Institute), and Virginijus Šikšnys (Vilnius University, Lithuania). Contrary to popular belief, Šikšnys was the first of the three to file a patent application covering his variation on the technology. Šikšnys filed for his patent on March 20, 2012,4 although it did not issue—that is, it was not formally granted by the U.S. Patent and Trademark Office (U.S.P.T.O.)—until May 2, 2017.5 For a number of reasons, Šikšnys’ contributions—both to the science CRISPR and the technology's patent dispute—have been overlooked.6 But Šikšnys’ patent covers using CRISPR-Cas9 through an in vitro, pre-assembled Cas9: RNA complex—also known as CRISPR ribonucleoproteins—an important iteration of the technology for a variety of applications.7 More famously, Doudna and Charpentier first filed their fundamental patent application covering CRISPR-Cas9 on May 25, 2012.8 Their original patent application contained over 150 claims—particular ways to practice the invention that defined the application's boundaries—and was notably unspecific with respect to cell type.9 Nonetheless, Doudna and Charpentier's patent attorneys pegged their clients’ invention broadly, as the use of a single-guide RNA to mediate the editing of genomic DNA.10 It was the ease, flexibility, and precision of this advance that has largely thrust CRISPR into the lay lexicon.11 During the pendency of Doudna and Charpentier's application in the USA, Zhang also filed a U.S. patent application—directed specifically to eukaryotic applications of CRISPR-Cas9.12 The principal improvement of Zhang's methods over his predecessors was the use of a nuclear localization signal and, separately, codon optimization to natively express Cas9.13 But Zhang's attorneys fast tracked his application through the U.S.P.T.O., a relatively expensive and strategically risky process.14 As a consequence, Zhang's patent—even though it was filed after applications from Doudna, Charpentier, and Šikšnys—was issued first in the USA.15 That quandary gave rise to the now-famous patent dispute in the USA, the first round of which was won by Zhang.16 The remainder of it is still being appealed; a decision is expected in late 2018.17 The U.S. interference decision, however, stands apart from the rest of the world. As detailed—excellently—by Knut Jørgen Egelie and his colleagues at Norwegian University of Science and Technology—the global CRISPR patent landscape is varied.18 Europe has now officially sided with Doudna and Charpentier over Zhang, although opposition proceedings at the European Patent Office have just begun.19 And China, too, recently sided with Doudna and Charpentier.20 These conflicting decisions are further complicated by a set of interlocking license agreements from the inventors’ biotech companies, with a great deal of uncertainty playing out in the global commercial sector for CRISPR.21 Unraveling those agreements, and the issues raised by the patenting of the technology in the first instance, speaks volumes about the values and pitfalls of patents in the research enterprise. One notable aspect of the CRISPR patent dispute is that it is, by and large, a dispute between academic research institutions. It pits lawyers representing the University of California against lawyers representing the Broad Institute of MIT and Harvard.22 To be sure, university rivalries are common.23 But because universities share among themselves a larger mission to create and disseminate knowledge to the public, litigiousness among them has been historically rare.24 University-against-university patent disputes, like CRISPR, complicate interinstitutional research agreements on several levels. First, they have the potential to chill formal interinstitutional research collaborations among universities if the institutions cannot agree on intellectual property issues beforehand.25 Universities may simply be unwilling to enter into such agreements in the first instance, or, perhaps more perniciously, discourage their faculty from informally developing such networks.26 While the empirical evidence for such a diminishment in collaborative efforts is slight—difficult to demonstrate, in part, because it requires the proof of opportunities not taken by universities—some recent survey data have found that ‘institutionally mandated [materials transfer agreements] put sand in the wheels of a lively system of intra-disciplinary exchanges of research tools’.27 Aside from this, there is substantial anecdotal evidence of institutional difficulties in creating such agreements.28 It stands to reason that, at least in some instances, these difficulties have ended some collaborations before they could begin. More immediately, this is a current issue with the CRISPR patent dispute given some internal dissention between Doudna and Charpentier's respective institutions concerning the intellectual property involved. Although Doudna and Charpentier filed their joint patent application in 2012, their institutions did not formally assent to a cross-licensing agreement until December 2016.29 If assenting to a cross-licensing agreement for a single piece of technology has proved difficult, it is unclear how the two institutions will deal with one another on future collaborations. Second, even with some friction between universities over obtaining patents for their researchers’ work, it has been rare for universities to sue one another regarding inventorship—until now. In 2011, for instance, the University of Utah sued the Max-Planck Institute concerning inventorship over a foundational group of patents concerning RNA interference technology.30 And since 2012, Stanford University and the Chinese University of Hong Kong have battled one another over lucrative patent rights to noninvasive prenatal genetic diagnostics.31 That dispute—despite several rounds of appeals—is still ongoing.32 Such patent disputes are costly, high stakes, and high profile. And while the CRISPR patent dispute itself is not a cause of such conflict, it has become emblematic—and potentially prophetic—of the tenor of such disputes today. Avoiding them in the first instance is a sensible institutional priority. But that sometimes comes at the cost of avoiding one's colleagues.33 Third, even apart from the administrative institutional level, patent disputes like these damper the culture of scientific collaboration, clearly something of tremendous import to modern science.34 Putting a price on a loosely defined culture of scientific collaboration is difficult—its loss is difficult to quantify. Nonetheless, many of the most significant breakthroughs of the past century arose in part from a culture of scientific openness and collegiality.35 Abandoning that in favor of inuring patent rights to researchers from a single institution seems, at best, unwise. Relatedly, it may erode scientists’ penchant for honest, if critical assessments, of their own work among collaborators and colleagues. A key piece of evidence used in the U.S. CRISPR patent interference against the University of California was a single one of Doudna's public statements that her collaborators ‘weren’t sure if CRISPR/Cas9 would work in eukaryotes—plant and animal cells’.36 That statement has now echoed throughout laboratories across the USA as a cautionary tale against critical reflections of one's work—at least while patents are pending.37 Lastly, patent conflicts’ hindrance of interinstitutional collaborations may simply be costly. Today, some research benefits from economies of scale, such as where expensive equipment can be shared among institutions.38 The New York Genome Center, for example, is a joint venture among several New York-area research institutions: NYU, Columbia, Cold Spring Harbor Laboratories, to name a few.39 This arrangement allows researchers at these institutions to share a fleet of Illumina X Ten sequencers, the total cost of which—including operations—runs into the millions of US dollars.40 Where research funding is diminishing—as is sadly the case in much of the Anglophone world41—universities may foolishly hesitate to engage in similar cost-saving arrangements in the short-sighted hope of avoiding future patent lawsuits.42 One would hope that the CRISPR patent dispute teaches others that such myopia isn’t warranted. Interinstitutional tensions aside, the CRISPR patent dispute raises some significant issues concerning patent licensing and commercialization—agreements between universities and commercial entities over the use and development of CRISPR. In CRISPR’s case, both the Broad Institute and the University of California have employed a system of ‘surrogate licensing’: ‘outsourc[ing] the licensing and commercialization of a valuable patent portfolio to a private company’.43 It is that company—rather than university—that takes responsibility for licensing the included patents to commercial researchers, including biotech startups and large pharmaceutical developers.44 At the same time, the surrogate is frequently working to develop the technology itself.45 This is certainly true for CRISPR. The University of California has delegated the entirety of its licensing rights to Doudna's inventions to Caribou Biosciences, which in turn has granted an exclusive license to develop human therapies to Intellia Therapeutics.46 The Broad Institute, meanwhile, has employed Editas Medicine as its surrogate for human therapeutics; the institute retains control over non-commercial and non-human therapy uses.47 Surrogates, therefore, ‘control a large and lucrative field for the exploitation of the licensed technology, and have significant freedom both to exploit it themselves and seek partners and sublicenses’.48 This system of surrogate licensing—while not unique to CRISPR—sets up several obvious conflicts. Surrogates may very well be unwilling to sublicense their technology to smaller biotech companies—who, in a very real sense, are rivals to the surrogate.49 Smaller companies seeking to develop similar uses of CRISPR to that studied by, say, Editas are unlikely to receive patent licenses to do so—at least on favorable terms.50 Surrogates are also not invested with the same public duty as their related academic institutions. Their duties, especially if they are publicly traded companies—as are Editas, Intellia, and Emmanuelle Charpentier's own company, CRISPR Therapeutics—are to their shareholders.51 In both real and legal terms, this duty tacks toward profit maximization rather than, say, advancing scientific knowledge or public access to the downstream products of their research—ideals typically lauded by research institutions.52 With respect to this conflict between public-facing goals and shareholder value, Michael Eisner, former CEO of Disney, put it best: ‘We have no obligation to make history. We have no obligation to make art. We have no obligation to make a statement. To make money is our only objective’.53 Lastly, surrogate licensing—even when functioning well—may ‘bottleneck’ the commercial development of the underlying technology.54 Surrogates may grant exclusive sublicenses that are too broad relative to their licensees’ contributions; this blocks others from developing competing technologies.55 Surrogates may also grant licenses to disease indications or areas of the genome far greater than any sublicensee can work at any given time.56 To be sure, bottlenecking is a serious problem with respect to university licensing as well.57 But universities are frequently more invested in nonexclusive licenses to commercial developers than for-profit surrogates.58 Most of the commentary on the CRISPR patents has been negative—and, in particular, the negative side of patenting the products of academic research.59 But—aside from money—there are some significant social positives as well. At their core, patents are rights to exclude others from practicing the claimed invention.60 The corollary to this axiom is that patents therefore allow their owners to dictate to the rest of the world how to use the inventors’ technology.61 This power to direct others’ research can be harnessed for societal good.62 Where the claimed technology raises ethical or social concerns, patent holders have the right to tell their technologies’ users to behave ethically and to provide access to downstream inventions.63 In this sense, patents—when used well—can function as a powerful form of private governance.64 This is certainly the case with CRISPR, the ethical and social issues of which have been explored at length.65 One potentially problematic use of CRISPR is its use in ‘gene drives’, a daisy chain of genetic editing that essentially forces future generations to inherit and subsequently pass on only a single variant of a particular gene.66 The concern, as detailed by Kevin Esvelt, is that gene drives, because they are forcibly heritable, become difficult to control once put in place.67 Should later research find negative, unintended effects of the particular genetic variant driven through the population, it may simply be too late.68 To that end, Esvelt and others have proposed patenting the use of CRISPR-based gene drives to, essentially, prevent others from using the technology without rigorous scientific and ethical controls.69 The legal mechanics of enforcing patent protection in this manner leave some gaps that likely need to be addressed. But Esvelt's proposal suggests, at a minimum, that patenting controversial technologies is one possible tool to further their ethical use. In other cases, rather than using patents to ethically restrict access to controversial technologies, patents can be used to ethically promote access to the same. That is, patent holders can demand licensees promise that they make their technology available to broad segments of society, and on fair terms.70 This is largely the case with Monsanto's license from the Broad Institute covering the use of CRISPR-Cas9 for a variety of agricultural purposes. That license essentially requires Monsanto to allow its farmer customers to save and resew seed from one season to the next, in contrast to some of Monsanto's past practices.71 Requiring this of Monsanto provides greater access to the fruits of CRISPR technology to farmers, who would otherwise be required to purchase expensive new seed each year from Monsanto.72 In the therapeutics context, similar license restrictions could be used, in theory, to require price controls, access plans, or that research and development funds be used, in part, to develop treatments for neglected diseases.73 And, perhaps counterintuitively, patents could also be used to ensure research access to a variety of technologies. Patent holders can publicly commit to refuse to enforce their patents against researchers or academic institutions. In the USA, these frequently take the form of ‘patent pledges’—‘commitments made voluntarily by patent holders to limit the enforcement or other exploitation of their patents’.74 Doing so both prevents others from patenting—and suing others—on the same technology, and dissuades less ethically minded competitors from entering the field.75 Patent holders can also use open licensing to researchers in developing and the technology for the public In the CRISPR context, this non-commercial use is through a a that provides access to CRISPR and through a patent licenses for academic use of the underlying To be sure, these restrictions have the potential for One ethical is to research. The for example, many patents to human methods of use and and therapies potentially from their But public over the a to restrict funding covering the has restrictions on its patent licenses concerning their technology's use in These restrictions have some among the scientific many of the not as an ethical but an of research for too, the on patents as the ethical uses of technology may out other less of of are not the only means of private governance to in ethically technology. The a research for patents as a tool for ethical governance in favor of transfer the The itself in ethical which or While the of the agreement is it stands to the of private ethical governance of technologies of patent In any the contrast among the and the that, like CRISPR patents are that can be used for or for At a minimum, ethically patent the of using patents as a tool for the public In many the ethical, legal, and social issues of CRISPR patenting are It is not that a genetic technology is with import to human and social And it is perhaps still that such an important technology the of a patent dispute among some of the research institutions. that the CRISPR patent dispute is a unique are some greater to be about the ethical, legal, and social of intellectual property in research The and perhaps most important for scientific is that and not research collaborations. and in particular, is largely a seeking to make the most significant in their turn to others at the of their for In this is perhaps by the recent of collaboration between and the of which have been in the CRISPR that the two made their through collaborative patent disputes research Doudna with and Zhang with of CRISPR research has now largely become in a of global and interinstitutional patent It is that further could be made without such Patent not as collaborative to be is the potential power of just the use and of their have some significant in how their institutions can use their who receive a license to the and set for academic are frequently the or of companies to their institutions out patent Doudna, for example, is the of Caribou Biosciences, the University of patent Charpentier, CRISPR and Zhang, Editas researchers with academic therefore have some control in how their technology will be with otherwise to the greater this they engage with and with their institutions to develop the fruits of their not these to university or their Lastly, the ethical, legal, and social of the CRISPR patents have something to about academic in a great deal of the academic on IP patents with a are patents are More of these them in of But th

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 68 citations exist upstream.