In 2010, Rebecca Skloot published The Immortal Life of Henrietta Lacks, a best-selling book about scientific ethics. By now the story is well known. Henrietta Lacks was an African-American woman who died of cervical cancer in 1951. Without her knowledge or consent, doctors took a sample of her cancer while she was still alive. That sample became the HeLa cell line, which almost immediately played a vital role in the development of the polio vaccine, and remains an important tool in medical and genetic research today. Scientific knowledge has advanced significantly over that period, and so have scientific ethics: Regulations introduced in the 1970s prevent researchers from subjecting people to experiments or studies without their consent, and today there is universal agreement that secret human experimentation is inexcusable.
The Lacks case remains compelling because it is not simply a historical episode; regulating human genetic research is a significant and ongoing problem. Today, donors give their genetic material knowingly, but they have very little control over the way their donation is used, and often cannot withdraw their donation from use once it is volunteered. Distributing benefits fairly is also a challenge: New genetic knowledge is often patented, which may reap profits for a firm at the expense of scientific advancement and the public interest. The benefits and the risks of human genetic research are both likely to grow over time. Fundamental research into disease, gene therapies, and personalized medicine may yield dramatic improvements in health care. But the continued expansion of genetic research may put the privacy and interests of donors in jeopardy—and existing policy is not equal to this challenge.
In 2013, federal officials made three major decisions about the regulation of human genetic research. First, the National Institutes of Health (NIH) admitted that the ethical issues posed by genetic sequences from the HeLa cell line—a line that now implicated descendants of Lacks without their consent—were too difficult for administrators to resolve on their own. The NIH therefore gave Lacks’s family strong consultative power over the uses of this material in research. Second, the Food and Drug Administration ordered a genetic-testing company called 23andMe to stop selling its home genetic-testing kits on the grounds that the tests amounted to unlicensed medical diagnosis. Finally, the Supreme Court issued a ruling on gene patenting that left scientists and lawyers puzzled about its logical reasoning and practical consequences.
Though most would agree that human genetic research is of expansive moral and practical importance, each of these decisions hewed to narrow technical considerations. Today there is no coherent regulatory framework for human genetic research in the United States, only a tangle of good intentions, agency guidelines, and judicial minutiae produced by decades of ad hoc policy-making.
Current regulations are problematic in at least three important ways. First, they do not provide adequate protection to donors of genetic material used in research, and they make it almost impossible for donors to withdraw from research already in progress. This poses a threat to the privacy of individual donors and allows scientists to conduct research that donors may find objectionable.
Second, courts regulate human genetic research through patent law. This obstructs research because the threat of lawsuits deters researchers from studying topics that involve patented genes. It also forces judges to make decisions based on their understanding of the state of the art in biotechnology. Unsurprisingly, given the complexity of genetic science, judges often misunderstand the issues at hand.
Third, U.S. regulations protecting donors and scientific innovations do not fit within a body of established international law. As a consequence, donors lack a clear legal means to protect their interests, and intellectual property protections granted in the United States are often not acknowledged abroad.
But there may be a viable alternative to the current regulatory patchwork: copyright law. Regulating genetic research through copyright law rather than patent law could allow donors to exert greater control over how researchers use their genetic material, while also preventing frivolous and obstructive patenting of the human genome. Though copyright law is evolving rapidly, nearly all countries subscribe to the same set of treaties, creating a comparatively uniform body of international law for both donors and researchers. Copyright may seem like an odd means of regulating science, but it more accurately reflects the genomic era of genetic research, in which researchers use computationally intensive methods to study large gene sequences. Aside from questions of practicability, this proposal may provide a way of refocusing the policy discussion on some of the deeper questions posed by human genetic research: Who possesses genes, what does this possession mean, and who should enjoy the rewards?
Copyright and Patent Law
Copyright and patent laws exist to protect and reward people who produce innovative ideas. They differ markedly in the kinds of protections they afford and the kinds of innovations they recognize. Both bodies of law are evolving, but copyright is more stable and internationally uniform.
Copyright law protects information that can be “fixed in a tangible medium.” It commonly applies to creative products such as writing, movies, or the contents of broadcasts. Copyright is granted automatically at the moment any eligible work is produced, and lasts from that moment until 70 years after the death of the creator. What is commonly termed copyright is actually a group of rights concerning reproduction, sale, claims to authorship, translation, and so on. This bundle of rights is flexible, and can be retained or transferred in many different ways.
Patent law, by contrast, protects inventions. In the United States, patents last 20 years and give patent holders the exclusive right to make use of the invention. Patents are not automatically given to inventions. Inventors must submit an application that demonstrates that the invention is eligible for patenting, and is novel, useful, and nonobvious. Technological advances have made it difficult to define what qualifies for protection. Software could be considered a copyrightable information sequence, or it could be a patentable way of usefully modifying the physical state of computer hardware. A maple tree is not a novel invention, but a genetically modified maple tree might be.
To put it kindly, patent law in the United States has become byzantine, and yields results that do not make intuitive sense. This is particularly so in the case of genetic research. The Bayh-Dole Act, passed in 1980, allows universities and similar organizations to seek patents on inventions produced with public funds. The U.S. Patent and Trademark Office has granted patents on human genes for some 30 years. As a consequence, a significant (but unknown) portion of the human genome has been patented, including genes that could substantially improve treatment of cancer and many other diseases. These developments suggest that a new approach to genetic research is badly needed. Copyright law, for all its complications, offers that alternative.
Three Problems for Human Genetic Research
The human genome has a moral gravity that exceeds the technical considerations of research or intellectual property law. Many people might prefer that human genes not be owned at all. Some would instead institute bans on many kinds of human genetic research, while others might treat the human genome as a common heritage of humankind, a legal status that similarly protects outer space and the open ocean. However, given the current regulatory hash and the advanced state of human genetic research, there is no obvious way of starting over with an entirely new approach. In practice, human genes are owned and very likely will continue to be owned. The practical concern is who will own them, and under what terms.
The American solution to this problem is unsatisfying: Human genes belong to whoever will claim to own them; or, having entered into general scientific knowledge, they do not belong to anybody and nobody governs their use. One alternative is for the state to assert a broad interest in controlling genetic research. These regulatory claims may be an effort to deter privatization: Brazil’s Genetic Heritage Management Council is charged with regulating research access to the country’s biodiversity. In effect, the government asserts that it holds partial rights over all of the genetic material to be found within the national borders. The claims may also be positive: Mexico has consolidated human genetic research under INMEGEN, a state-run research institute with a broad mission to describe the genetic makeup of the country’s population and use this knowledge to improve medical care. The massive UK Biobank relies upon an idea of shared ownership and responsibility. These approaches are relatively new and represent national forms of ownership, and their success is an open question. For a number of reasons, such approaches are not viable for the United States, legally or politically. In practice, implementing this kind of regulation would mean nationalizing the biotechnology industry. This would probably be unconstitutional, and would certainly meet opposition from left, right, and center.
And so we are stuck with our current regulatory regime. As outlined above, there are three major problems with the American system. First, donors of genetic material are not sufficiently protected. Second, gene patents obstruct scientific access to important material. Third, neither of these approaches squares with practices in other countries. Let’s examine each of these.
Patents on genes, tests, or drugs come at the end of a long process of research. The process begins with people who donate their DNA to science. Some donors give their material as part of research on a specific disease or problem. Others give to projects that make material available for all kinds of research. The motives of donors are generally altruistic: They are not significantly compensated, and often do not stand to benefit personally from new discoveries.
The entire process affords the same protections given to medical patients. Under this model, proposed research projects are evaluated by local expert panels known as institutional review boards (IRBs). In assessing the ethical issues raised by a project, IRBs pay particular attention to the possible benefits and risks to participants. The autonomy of patients is the foundation of this approach to research ethics. Patients must be free in their decision to participate, and do so only after being presented with the information needed to make an informed, deliberate choice.
A person who intends to donate genetic material receives a description of immediate risks (minimal) and benefits (also minimal—donors ordinarily surrender any right to the future benefits of research). Prospective donors also receive a statement of the intended purpose of the research, a description that is usually calculated in its vagueness. In fact, donors are routinely asked to give “blanket consent,” allowing their samples to be employed for virtually any legitimate scientific purpose, forever, and they usually give it.
The most obvious risk for participants in genetic research is a loss of privacy. For this reason, donor samples, both physical cells and digitized gene sequences, are anonymized. Anonymity is certainly important, as donated genetic information is commonly published in research databases that are accessible by anybody who cares to look at them anywhere in the world. But it is not clear that anonymization provides real security now, nor that it will continue to do so in the future. The threats to anonymity grow as more researchers employ donated data. In 2013, a team of researchers led by Melissa Gymrek called attention to this problem by de-anonymizing the samples of several Americans whose full gene sequences had been published in open-access databases. Identifying the samples did not require any hacking, breach of ethics, or sleight of hand; the team only needed publicly available information from genealogy websites.
To be sure, the possible harms of a breach of confidentiality, at least in the United States, have been lessened by recent laws. The 2008 Genetic Information Nondiscrimination Act makes it illegal to use a person’s genome to determine decisions about employment or most kinds of insurance, while the Affordable Care Act guarantees access to insurance for people with existing medical conditions. However, it really isn’t possible to anticipate all the real-world consequences of a breach of privacy. As science understands more about the human genome, more personal information may well be revealed. For this reason, donors might reasonably volunteer their genes now, and might just as reasonably want them withdrawn decades hence. Blanket consent makes this impossible. What we need to preserve is a principle of autonomy in light of the uncertainty that still exists about human genetics. We can no more anticipate the state of genetic knowledge 60 years from now than the doctors who created the HeLa cell line could have anticipated the uses to which these cells are put today.
The medical approach to protection is also troublesome because genetic research reveals information about people other than the individual donor. A donor’s relatives and descendants share many of the same genes. This was a major concern for the Lacks family, and the main reason why the NIH gave her family some power over how these cells are used. The NIH has emphasized that this is not meant to set precedent, only to redress a previous injustice and its long afterlife. But maybe it should set a precedent: Many of the concerns about HeLa cells also hold for human genetic research generally. It is not possible to foresee whether existing samples will remain anonymous, nor is it possible to know what kinds of knowledge science will be able to produce from gene sequences in the future. Past donors, if given a choice, might prefer that their genetic material be withdrawn from research after a certain time, or when new technologies are developed that did not exist at the time of donation. They also might wish to impose reasonable restrictions on its use.
To a lesser extent, genes are also shared by populations and ethnic groups. Material donated with broad consent may also be used for research that is objectionable to such groups. The recent case of the Havasupai Indians in Arizona calls attention to the gap between what is technically permitted by many consent forms and what is socially appropriate and responsible. In the early 1990s, the Havasupai donated genetic material for what they understood to be a study of diabetes. However, the documentation allowed researchers to use the samples for many other purposes, including studies of inbreeding, schizophrenia, and the tribe’s geographic origins. The geographic study directly contradicted the tribe’s religious beliefs about its origin. The other studies, they felt, implied that the tribe was inbred and plagued by severe mental illness.
Scientists could avoid pitfalls like this by remaining in closer contact with the groups they study, and listening to their views about what kinds of research are appropriate. But existing regulations make this sort of dialogue difficult. First, donors often have no way to compel researchers to listen to their concerns, because they cannot withdraw. Second, privacy protections often make it impossible for researchers to follow up with donors.
In practice, regulation is mostly in the hands of scientists and companies. The 2003 Fort Lauderdale Agreement—the major set of guidelines for the sharing of human genetic data—divides responsibility for regulation among funding agencies, data producers, and data users; there is no role designated for donors or advocates. Similar views are expressed in a 2008 journal article that offered the consensus view of leading bioethicists; as a matter of principle, the bioethicists believe that donors should be able to withdraw from research, but because there is no practical mechanism for doing so, donors should be told that their consent is effectively irrevocable. The lack of a practical mechanism is the most significant problem. Though real, willful violations of research ethics do occur, many more problems arise by accident, either through predictive injuries, misuse of scientific claims by nonscientific communities, or simple ignorance about the wishes of donors. The scientific community is in many ways highly effective at self-regulation, but at present it engages in regulation without a full awareness of the participant and public interests at stake.
Questions about the regulation of research lead directly into questions about who should benefit from innovations and how these innovations should be legally protected. To date, developments in human genetic research have been protected under patent law. In the case decided by the Supreme Court in 2013, Myriad Genetics, Inc. had patented two genes—BRCA1 and BRCA2—whose variants can be predictive of breast and ovarian cancer. These patents protected the company’s highly profitable genetic test, BRACAnalysis. However, the BRCA gene sequences themselves are widely known; in fact, they can be freely downloaded from a number of databases. As important predictors for major cancers, they are employed by researchers in many fields.
The importance of open scientific access to the BRCA genes did not factor in the Court’s decision: The legal case turned on the technical details of patent law. The Court unanimously held that DNA sequenced from a biological sample is not patentable, because such a gene sequence is a product of nature. However, if the sequence is synthesized in a lab it is not natural, and is therefore entitled to patent protection, even though the natural and synthetic sequences have the same content and function. Technology for synthesizing DNA is a standard research tool, so the requirement that a sequence be produced synthetically to be patentable is in fact no obstacle. A decision intended to limit the scope of gene patenting, therefore, may produce the paradoxical effect of extending patent protection.
Patents are not the right way to protect scientific innovations in genetic research. The legal protections afforded to patents are much broader than the protections required for many research or business purposes. Because they are so broad, they obstruct scientific progress. A 2003 study by Mildred Cho and her colleagues found that more than half of major research labs let patent concerns affect their research decisions. The mere threat of infringement deters researchers from studying important problems, leading many to describe gene patents as creating an “anticommons,” in which a shareable resource is needlessly and counterproductively divided up among a large number of competing rights holders. Although Myriad Genetics patented the BRCA genes to protect a diagnostic test, the patent also had the effect of making research on an important biological discovery technically subject to lawsuits.
These patents are also bad for medicine. The patent that gave Myriad a monopoly on BRCA testing is analogous to giving a patent on using an X-ray to observe a broken bone. Exclusivity made the test prohibitively expensive for many women who might otherwise have learned whether they faced a serious risk of breast and ovarian cancer. The lack of competition also allowed Myriad to offer a test that was less accurate than it should have been. There is a reason that the United States has no exclusive purveyor of X-ray diagnosis: Tests that simply observe a correlation between a physical state and an outcome are a law of nature, and laws of nature have never been eligible for patent. In the case of genes, judges have struggled to define where nature ends.
In 2002, legal scholar Andrew Chin created a very inexpensive proof of concept intended to demonstrate the perversities of gene-patent law. Under patent law, pre-existing scientific knowledge of the gene sequence in question, or “prior art,” is grounds for denying a patent. This prior knowledge is mostly unpublished, because it is of no scientific value to publish enormous catalogs of identified gene sequences of arbitrary lengths. Chin, noting this problem, produced a CD containing a brief description of standard practices in genetic laboratories along with a large catalog of DNA sequences. This single disc, which cost no more than a dollar to create and recorded information of no practical utility for researchers, has been used to challenge dozens of gene patents over the last decade (though it did not play a role in the Myriad case).
Chin’s CD also underscored an important point. The scientific study of genes is much more like reading printed information than performing physical manipulation. A great deal of important genetic research doesn’t take place in a stereotypical lab filled with test tubes, scientists in lab coats, and expensive equipment. The research happens on computers, and involves analysis of large quantities of digital, textual representations of complete human gene sequences; this big-data approach to research is commonly called genomics. The amount of information employed in genomic research is vast—thousands of complete human genomes. The expectation, as yet unrealized, is that the practical knowledge to be derived from this information is comparably vast. For many researchers, the ultimate goal is personalized medicine, in which health care would be tailored to fit the genetic makeup of the individual patient.
Research on this scale is necessarily international. Donors for the 1000 Genomes Project come from 17 countries. Scientists collaborate across borders. The discoveries have practical consequences for people throughout the world. However, international law on both research subjects and intellectual property is limited and contradictory.
The United States and European Union both attempt to provide strong protections for donors, but do so by very different means. Other countries do little to protect research subjects. India, which combines a strong research infrastructure with weak protections for human subjects, has become an attractive site for biomedical research, some of it less than scrupulous. Uneven standards of research ethics may trigger a race to the bottom.
The lack of legal harmonization also creates tangles for transnational research projects, which may collect genetic samples in one country, sequence and store those samples in another, and analyze them throughout the world. Different bodies of domestic law do not agree about basic questions such as the patentability of human genes. Nor do they agree on the appropriate grounds for making such decisions. Because it is a matter of life and death, many poor countries ignore biomedical patent claims, or have honored them only after the threat of crushing sanctions. In Europe, human rights law has played a significant role in the development of gene-patent law, which is consequently narrow in its protections and includes important exceptions for the public interest. In the United States, moral questions have been held wholly to the side, with legal cases instead focusing on technicalities of the kind discussed here.
A New Approach to Genetic Research: Copyright Law
As these problems suggest, our current regulatory regime for genetic research is woefully inadequate. Copyright law offers a promising alternative, not only to improve our ability to protect donors but also to make sure that genetic material isn’t closed off from further research.
Such a shift isn’t so outlandish. Science has already begun to handle genetic information like text; as information scientist Amelia Acker has shown, biological samples are standardized and cataloged in much the same way as text archives. Indeed, an entire scientific field, bioinformatics, exists to organize and study biological data. On its face, it is possible to view a genetic sequence as copyrightable: It can be “fixed in a tangible medium”—in fact, in many different tangible media—and this information can be readily reproduced and disseminated.
But can copyright law actually solve the three problems discussed above concerning donor privacy, innovation, and international coherence? To be sure, copyright law has been maligned for many of the same reasons as gene patents: Copyright is used to enclose material that should be freely available, can obstruct the movement of important knowledge, and deters creative uses of information. However, copyright also offers a highly flexible set of rights and has recognized international validity. It has also been successfully employed to promote openness. A particularly notable success is the General Public License, the legal basis for the development of open-source software. A copyright approach to human genetic material could allay our concerns with the current system—and, most importantly, provoke a shift in our thinking about human genetics and public policy.
Here’s how it could work: When people donate their genetic material for research, they could assert a copyright on the genetic information that is produced when this material is sequenced. Donors, having asserted copyright on their genetic information, could license it for use. Such a license need not require monetary compensation—in fact, existing ethics rules would bar researchers from paying donors a significant price for their genetic material. And such a license would be granted for only a fixed time period. This would allow donors to withdraw their information from use if they no longer wished to be research participants or felt that their donation has been misused. Because copyright survives its original owner, the families of donors could retain decision-making ability of the kind extended to the relatives of Henrietta Lacks. Because licenses would have to be renewed, donors, researchers, and the public would need to engage in regular deliberation about the state and trajectory of human genetic research. Such discussion is a public good in itself. Now, granted, a copyright approach would place some limits on anonymity: At minimum, some durable (though nonpublic) record would need to exist to connect the sequence with the donor. But donors with copyright would have access to legal protections and remedies for breaches of privacy that they do not have currently. And some degree of identifiability could also allow donors and researchers to engage in sustained dialogue over the use of genetic information.
Information licensing of this kind could also check the second problem I identified, which is the enclosure of the genetic commons made possible by patent law. The assertion of copyright would preempt subsequent patent claims in much the same way as Andrew Chin’s CD. Under a copyright regime, Myriad Genetics never could have patented BRCA1 and BRCA2.
Some would object to such a system. Believers in a pure public domain, particularly numerous in the sciences, might object that this form of licensure would unduly limit access to information. But would an unconstrained public domain of human genetic information really be desirable? People who possess certain genetic sequences have a legitimate interest in regulating how these are used—and donors are uniquely equipped to identify their own interests. A public domain requires a public to manage it responsibly.
Meanwhile, others defend the current gene-patent system because they believe as a matter of principle that market forces are the most important spur to research. Many firms also benefit financially from gene patents and defend them out of self-interest. Both groups would surely object to the copyright approach, just as they have long objected to the claim that gene patents create an anticommons. However, extending premature legal protection to research is a much greater threat to innovation. Gene patents slow down or halt important lines of research solely to protect detective or descriptive tests that do not improve scientific knowledge or medical care. Genetic information licensing could deter precisely this kind of unnecessary privatization of shared knowledge.
A copyright approach also solves the third problem I discuss: the lack of legal harmonization across borders when it comes to genetic research. Nearly all of the countries in the world are signatories to a number of copyright treaties overseen by the World Intellectual Property Organization, a UN agency. Though enforcement of copyright, in practice, varies greatly from country to country, the treaties provide a relatively uniform legal framework. This could allow donors to assert their rights even when research crosses national borders. To be sure, enforcement of copyright and access to courts vary a great deal from country to country, but creating rights that are enforceable in principle would be a large improvement over the present situation, where donors often have no legal recourse.
There is a major practical problem in all of this: Generally speaking, legally enforcing copyright is expensive and time-consuming. But that need not be the case with genetic research. For one thing, there are only a few key sources of digitized biological data. And the number of research organizations that can make practical use of openly accessible genetic information is also quite small. Though I can download thousands of complete human genomes, I do not have any of the scientific knowledge or the supercomputer or gene-synthesis lab to do anything with it that would be commercially valuable or personally injurious. In other words, meaningfully infringing a copyright on a genetic sequence would be much harder, and involve higher upfront costs, than pirating a movie or plagiarizing an article. That, and the high degree of self-regulation in the scientific community, would make detection of inappropriate data use more likely.
Finally, one of the strongest recommendations for this approach is the low marginal cost of failure. If copyrights proved unenforceable, the result would be a weakly regulated public domain of the kind that already exists. If copyright came to be used as a legal instrument to enclose the genetic commons, or created conflicting rights claims, this would not be so different from the present confusion of gene-patent law.
Making Copyright for Genes Happen
Human genetic research is developing rapidly. Scarcely a decade after the completion of the Human Genome Project, the sequencing (and comparison) of complete human genomes has become a common and relatively inexpensive technique. Though new practical applications continue to emerge, research is in many ways becoming more abstract: Human genes are increasingly studied in digitized forms and treated as information rather than physical structures. This acceleration of research presents major challenges for the protection of genetic donors, the balancing of public and private interests in the fruits of research, and the responsible conduct of international research.
Federal regulation has lagged behind. It has often formulated policy by analogy to other issues. Just as importantly, it often fails to consider social and moral concerns that are unavoidably present in human genetic research. Copyright, while imperfect, might give donors more robust protections, while also preventing excessive or premature privatization of scientific innovations. Considering copyright as a framework for human genetics also offers a way to recast the problem as one of access to and control of information. In this respect, human genes do not pose a unique moral or policy problem, but typify a more general set of challenges that are intrinsic to our information society.
How exactly can we begin to change the existing regulatory framework? Regulators and researchers are aware of the challenges and eager to find solutions. Several actors could try out copyright licensing. The Office for Human Research Protections (OHRP) protects the rights of participants in federally funded research. At the prompting of the Obama Administration, the OHRP has contemplated a number of changes to existing regulations, particularly those that involve human biological material. Copyright licensing could be entertained as another possible modification. The NIH could also try out the policy: Its binding consultation with the family of Henrietta Lacks is a practical, though limited, application of the principle. Any large research consortium could introduce copyright licensing for new donations. Such consortia have developed many innovative ways to protect donors and researchers, though these efforts have fallen short because they do not meet the question of ownership head-on.
In a narrowly legal sense, genetic copyright licensing is a judicial question. It is impossible to guess how courts might decide, but a form of licensing on the model of copyright could work in practice even if courts rejected a direct assertion of copyright. However, the issue at hand is not simply legal—it is also practical and moral. As human genetic research progresses, there is a pressing need to change regulations to allow gene donors to protect their own interests, and to assure that the benefits of science are not unfairly hoarded.