A full version of this article, published in the Fall 2009 issue of the Harvard Journal of Law & Technology, is available here.
Today, more than 300 million tissue samples from more than 178 million individuals are stored in the United States, and this number has been growing by more than 20 million samples every year. Through genetic analysis, researchers hope to identify disease-related and other genes and to measure the frequency of such genes’ occurrence across large populations. This kind of research requires population-wide bio-repositories of samples available for study.
Yet, individuals providing tissue for research may hesitate to do so if they fear that their interests will not be respected. Tissue providers may have concerns that their cells and genetic material — materials with which they may strongly self-identify — will be used for research they find morally repugnant or about which they were not informed. Unanticipated disclosure of genetic information may negatively impact the ability of unwitting tissue providers and their close genetic relatives to obtain insurance coverage or appropriate medical treatment. Tissue providers may also have strong interests concerning the commercialization of their cells and genetic material, especially if they are not permitted to share in the profits. And tissue providers have a further interest in the progress of science and medicine.
Researchers and society both also have strong interests in how tissue is used. Scientific research using human cells can be (and has been) immensely beneficial. Inappropriate or onerous restrictions on human tissue research may negatively impact the progress of science and medicine—a concern about a tragedy of the anticommons clearly articulated by several courts that have presided over disputes about the ownership and control of human tissues used in research. Moreover, permitting tissue providers to commercialize their cells may divert tissue from worthwhile research, diminish the necessary incentives for research funders to invest in research and development, or undermine societal dignitary interests.
Failing to mediate tensions between tissue providers, researchers, and society may cause individuals to avoid research participation. The solution is not, however, to compel such participation by dispensing with consent altogether and generally conscripting tissue left over from, for example, blood draws or biopsies. Indeed, doing so might cause individuals concerned about the future use of their genetic material to forego routine medical care in order to prevent their cells from being so conscripted. Rather, the interests of tissue providers, of researchers, and of broader society each demand respect and protection.
Given that tissue providers and researchers focus primarily on the importance of genetic information, one solution may be to recognize a limited right for individuals to control how the information contained in their cells is used: an informational property right. This approach draws on American intellectual property law. Both the proposed informational property right and intellectual property rights arise in part because they encourage innovation and investment in research. Patent and copyright encourage would-be inventors to invest time and resources in hopes of securing certain time-limited exclusive rights with respect to the result. Informational property rights would encourage investment in research in two senses: first, by enabling individuals to provide their tissues for research use with confidence that their interests will be respected (“investment” as resources); and second, by increasing public trust in the ethical conduct of scientific research (“investment” as both public resources and emotional energy).
American copyright law, in particular, has features from which a useful informational property model can be drawn. Unlike tangible property, copyright cannot be lost through unconscious abandonment during its statutory period, and its protections extend even to unpublished (i.e., undisclosed) works. Personal genetic information protected by a copyright-like informational property right would thus be unavailable for unauthorized use no matter how or from where it was obtained. Informational property controls would likewise persist even in downstream creations like digitized genetic sequences, as the information contained in such records would be identical to that of the original cells, even if their forms were different. Cell lines might also be subject to informational property controls where a tissue provider’s genetic sequence and the genetic sequence of a cell line are similar in material ways.
Under a system of personal informational property rights, any researcher wishing to gain access to the information contained in an individual’s cells would need to obtain a license from that individual. Such a license would fill much the same role that an informed consent document plays in our current system — a license authorizes access to and use of the desired material.
Critically, tissue provision should be “copyleft” in nature. Such licenses are “some rights reserved,” and downstream researchers may freely use, copy, modify, and distribute the protected material subject to specific restrictions enumerated in the license. In particular, a copyleft license often requires that downstream creators “distribute derivative works only under a license identical to the license” that governs the original work, here the tissue sample. This kind of “viral licensing” provision would reinforce tissue providers’ ability to exercise their interests with respect to derivative works, such as digitized DNA sequences, cell lines, or other commercial products.
Moreover, by standardizing the range of authorization that can be granted or withheld, and obtaining this authorization upfront, copyleft licensing assuages many of the anticommons concerns identified by courts and regulators declining to impose rigorous consent requirements on those obtaining tissue for research purposes. The Greenberg court, for example, declined to extend a duty of informed consent to cover economic interests because imposing such a duty “would give rise to a type of dead-hand control.” Rather than requiring each potential user to negotiate independently with the rights holder, copyleft licensing minimizes transaction costs while still preserving considerable provider control.
Turning to specifics, the interests of tissue providers signal the kinds of conditions that would be useful in a copyleft informational property licensing system. Creative Commons licensing, for example, permits combinations of four kinds of conditions limiting the use of protected material by downstream users: attribution (“BY”), noncommercial (“NC”), no derivatives (“ND”), and share alike (“SA”). Although the number and types of conditions might expand beyond those recognized by Creative Commons, the range of possible conditions could nonetheless be cabined and standardized. The provider’s interest in control, for instance, might be achieved through restrictions placed on the types of research for which a tissue sample may be used. Confidentiality interests could be addressed with license terms respecting the anonymization of tissue samples and the disclosure of identifying genetic information. And while the traditional Creative Commons noncommercial condition only enables creators to prohibit commercial uses of their works (and derivative products based on those works), licenses in this context could permit tissue providers to set a variety of different conditions regarding research-related commercialization. For instance, tissue providers could elect to prohibit commercialization — or at least restrictive patent enforcement — of derivative products outright or, alternatively, elect to require benefit-sharing of the fruits of those products.
In this manner, Creative Commons and other copyleft mechanisms evoke a tiered informed consent approach, which the National Cancer Institute has formally adopted in its best practices. Tiered consent documents present potential tissue providers with a menu of research categories to which they may consent. Potential tissue providers may consent to some, all, or none of the research categories presented, facilitating provider choice while constraining that choice to a manageable set of options.
Adopting a legal regime analogous to copyright for protecting personal genetic information is not without difficulties. For one, copyright’s fair use exceptions threaten to undermine the whole project of providing effective protection for the interests of tissue providers in research. Fair use permits unauthorized use of protected works for certain socially valuable purposes, including education and research. If such exceptions applied wholesale in the context of informational property rights, it is likely that some, if not all, human tissue research would be construed as fair use.
In this context, however, consider an alternative, fair use cognate permitting research aimed at discovering information about the regions of DNA that all persons have in common—more than 99% of human genetic sequences. This “fair utilization” exception might well encourage adoption of an informational property model and provide useful balance between tissue providers’ interests and the free flow of research materials. Fair utilization would permit generally unobjectionable research to proceed with fewer hurdles while maintaining adequate protection for the interests of tissue providers where such protection is desirable. None of the genetic information obtained pursuant to fair utilization would be unique to a single individual, and therefore confidentiality concerns would be minimized while concerns about control would not be significantly undermined.
If a line between shared and unique regions of human DNA can be drawn with sufficient precision, fair utilization would yield at least three advantages for the informational property approach. First, it would further limit the transaction costs of human tissue research by limiting the circumstances in which more detailed authorization is required. Second, incorporating fair utilization would begin to address the tensions that shared genetic identity present for traditional notions of private property. Fair utilization would open access to material for purposes of researching those portions of DNA that are shared among all humans, essentially creating a limited genetics common. Finally, fair utilization could bring commons-based and private-property approaches into harmony. By permitting individuals to assert greater control where individualizing genetic information may be at stake, the cabined nature of fair utilization provides more rigorous protection where tissue providers have a stronger stake in the research enterprise.
Integrating Informational Property with Existing Systems
The informational property model envisioned here provides a more complete approach for protecting the interests of tissue providers and satisfying the needs of researchers than other proposals focusing on tort, contract, or traditional property. In tort, for instance, informed consent doctrine has been largely unhelpful in protecting tissue providers. Although medical professionals who fail to obtain adequately informed consent may be subject to tort liability, whether non-physician researchers bear similar obligations is less clear. Courts have also generally disfavored non-physical or dignitary injuries, excluding precisely the kinds of harms sustained by tissue providers whose interests in control, confidentiality, and commercialization are not respected.
Contract doctrine proves even less helpful. The contract approach has been successfully deployed at least once: PXE International negotiated for co-ownership of any patents resulting from study of blood samples collected from PXE patients and their families. In most instances, however, a contract is likely to be inadequately protective because, among other things, it effectively puts the burden of initiating negotiation on the tissue providers, who will almost always be the less-informed and less-powerful bargaining partner.
Recognizing private property rights in human tissue—a property right in the physical cells obtained through tissue extraction—fares no better. Most critically, property principles provide little recourse for those who have abandoned their property. Hair and dead skin cells lost from the body in daily life are generally considered abandoned property. Cell samples obtained during routine interactions with medical or research personnel are classified as waste material once their clinical purpose is complete, and waste material is sure to be considered abandoned property. And cells that are considered abandoned property may be used without permission for any purpose, including privacy-invading ones.
A Multi-Model Proposal
While the informational property model is better suited to governing tissue provider-researcher interactions than tort, contract, or personal property, this model alone is unlikely to yield a complete governance regime. Informational property assigns an enduring property-like right and binds both direct and downstream users of tissue, but tort supplies most of the remaining foundation. Tort-imposed duties of care can give form and substance to what is required in obtaining informed consent and appropriate licensing, a role that existing doctrine already performs in the context of informed consent for medical treatment. Tort is also necessary for setting baselines for what privacy and confidentiality controls must protect—identifying what reasonable expectations of privacy should be and imposing liability on those researchers or other actors who fall short of these expectations.
Protecting the interests of tissue providers and researchers in human tissue research thus demands a mixed legal regime, drawing not only on recognized tort principles, but also on informational property rights—the seeds for which already exist, but which the current legal regime does not formally recognize.