Environmental Policy in the Age of Genetics
Rapidly emerging technology could lead to a fundamental shift in the way we safeguard human health.
In April 1965, a young researcher at Fairchild Semiconductor named Gordon Moore published an article in an obscure industry magazine entitled "Cramming More Components Onto Integrated Circuits." He predicted that the power of the silicon chip would double almost annually with a proportionate decrease in cost. Moore went on to become one of the founders of Intel, and his prediction, now known as Moore's Law, has become an accepted industry truism. Recently, Monsanto proposed a similar law for the area of biotechnology, which states that the amount of genetic information used in practical applications will double every year or two.
Sitting at the intersection of these two laws is a fascinating device known as the gene or DNA chip, a fusion of biology and semiconductor manufacturing technology. Like their microprocessor cousins, gene chips contain a dense grid or array placed on silicon using techniques such as photolithography. In the case of gene chips, however, this array holds DNA probes that form one half of the DNA double helix and can recognize and bind DNA from samples taken from people or organisms being tested. After binding, a laser activates fluorescent dyes attached to the DNA, and the patterns of fluorescence are analyzed to reveal mutations of interest or gene activity. All indicators are that the gene chips are obeying Moore's Law. Three years ago, the first gene chips held 20,000 DNA probes, last year the chips had 65,000, and chips with over 400,000 probes have recently been introduced. The chips are attracting intense commercial interest. In June 1998, Motorola, Packard Instrument, and the U.S. government's Argonne National Laboratory signed a multiyear agreement to develop the technologies required to mass-produce gene chips.
So what is new about this technology? Experimental chips are already at least 25 times faster than existing gene sequencing methods at decoding information. The chip decodes genetic information a paragraph or page at a time, rather than letter by letter, sequencing an entire genome in minutes and locating missing pieces or structural changes. If we can read a person's genetic story that fast, we can finish the book in a reasonable amount of time and understand more complex plots and subplots. Existing techniques have been valuable in identifying a small number of changes in the DNA chain commonly known as single nucleotide polymorphisms, which may result in diseases such as sickle cell anemia. However, these approaches have proved too slow and expensive to provide information on polygenic diseases, in which many genes may contribute to the emergence of disease or increased susceptibility to stressors. The gene chips are a key in recognizing this multigene "fingerprint," which may underlie diseases with complex etiologies involving the interaction of multiple genes as well as environmental factors.
Much environmental regulation protects human health by a very indirect route. For example, a very high dose of a chemical might be found to cause cancer in rats or other laboratory animals. Even though the mechanism by which the cancer is formed may be poorly understood, an estimate is made that a certain amount of that chemical would be harmful to humans. Estimates are then made about what concentration of that chemical in the environment might result in a high level in humans and what level of discharge of that chemical from an industrial plant or other source might result in the dangerously high concentration in the environment. Finally, the facility is told that it must limit its release of that chemical to a specific level, and, in many cases, the technologies to accomplish these reductions are prescribed. This long series of assumptions, calculations, and extrapolations makes the regulatory process slow, inexact, and contentious-a breeding ground for litigation, scientific disputes, and public confusion.
Gene chip technology could turn that system on its head. Biomarkers (substances produced by the body in response to chemicals) have already made it possible to measure the level of a specific chemical such as lead, benzene, or vinyl chloride in an individual's urine, blood, or tissue. Gene chips will make it possible to observe the actual loss of genetic function and predict susceptibility to change induced by a chemical. As the cost of the technology decreases, it will be possible to do this for many, many more people; ultimately, it might be cost-effective to screen large populations. The focus of environmental management will shift from monitoring the external environment to looking at how external exposures translate into diseases at a molecular level. This could radically change the way we approach environmental risk assessment and management, especially if diagnostic information from the gene chips is used in combination with emerging techniques in the field of molecular medicine. This could open up whole new avenues for prevention and early intervention and allow us to custom-design individual strategies to reduce or avoid a person's exposure to environmental threats at a molecular level. Some simple intervention measures already exist. For instance, potassium iodide can block a type of radiation that causes thyroid cancer, and the Nuclear Regulatory Commission has recently approved its distribution to residents living in close proximity to nuclear power plants. However, unlocking the Holy Grail of the human genome moves the intervention possibilities to a very different level. New techniques are now being developed that block the ability of environmental toxins to bind to proteins and cause damage, speed up the rate at which naturally occurring enzymes detoxify substances, or enhance the ability of the human body to actually repair environmentally damaged DNA. We move from the end-of-the-pipe world of the 1970s to the inside-the-gene world of the next millennium.
This potential comes packaged with significant dangers. Francis Collins, director of the Human Genome Project at the National Institutes of Health, recently remarked that the ability to identify individual susceptibility to illness "will empower people to take advantage of preventive strategies, but it could also be a nightmare of discriminatory information that could be used against people."
Without the proper safeguards in place, possibilities will abound for coercive monitoring, job discrimination, and violations of privacy. From a policy perspective, the danger exists that we could either overreact to these potential problems or react too late. Some of the more obvious issues are being addressed by a part of the Human Genome Project that looks at ethical, legal, and social implications of our expanding knowledge of genetics. However, the privacy and civil liberties debate has tended to mask more subtle, but potentially profound, effects on fields other than medicine. The use of gene chips could forever alter the rules of the game that have dominated environmental protection for 25 years. Here are a number of speculative concerns for those responsible for environmental policy.
First, as such testing and intervention capacity becomes cheaper, more accessible, and more widespread, it puts more power in the hands of the public and the medical profession and takes it away from the high priesthood of toxicologists and risk assessors in our regulatory institutions. This is not necessarily bad, because polls have shown that the public has a greater trust in the medical profession than in the environmental regulatory community. However, it is not at all clear that the medical community wants, or is trained, to take on this role. Research done by Neil Holtzman at the Johns Hopkins School of Medicine has shown that many physicians have a poor understanding of the probabilistic data generated by genetic testing, and other studies have indicated that many physicians are uncomfortable about sharing such information with patients. The few genetic tests already available for diseases such as cystic fibrosis have taxed our capability to provide the counseling needed to deal with patient fears and the new dilemmas of choice. Added to this picture is the potential involvement of the managed care and insurance industries in defining the testing, treatments, costs, and ultimate outcomes. Genetic information could be used by insurance companies to deny coverage to healthy people who have been identified as being susceptible to environmentally related diseases. Knowledge is power, and if the gene chips provide that knowledge to a new set of actors, environmental decisionmaking could be radically altered in ways that provide immense opportunity but that could also result in institutional paralysis, mass confusion, and public distrust.
Second, in a world where environmental policy is increasingly driven and shaped by constituencies, the new technologies offer a stepping stone toward the "individualization" of environmental protection and are a potential time bomb in our litigious culture. The rise of toxic tort litigation over the past 25 years has closely paralleled our scientific ability to show proximate causation; that is, to connect a specific act with a specific effect. Until now, environmental litigation has fallen largely into two classes: class action suits filed by large numbers of individuals exposed to proven carcinogens such as asbestos, or suits brought by people in cases where exposures to environmental agents have lead to identifiable clusters of diseases such as leukemia. The possibility that individuals could acquire enough genetic evidence to support lawsuits for environmental exposures raises some truly frightening prospects. Though workers' compensation laws generally bar lawsuits for damages resulting from injuries or illnesses in the workplace, loopholes exist, especially if employers learn of exposures and/or susceptibilities through genetic testing and do not notify workers. The expanded use of gene chips for medical surveillance in the workplace increases the possibilities for discrimination across the board. Finally, the testing of large populations with this technology may increase the likelihood of legal disputes based on emerging evidence of gender-, ethnicity-, or race-based variances in susceptibility to environmentally linked diseases. We are quickly wandering into an area with few legal protections and even fewer legal precedents in case law.
Third, the increased knowledge of human genetic variation and vulnerability will likely increase what Edward Tenner of Princeton University has described as the "burden of vigilance"-a need to continuously monitor at-risk individuals and environmental threats at levels far exceeding the capacities of our existing data-gathering systems. This could result in a demand for microlevel monitors for household or personal use, better labeling of products, and far greater scrutiny of the more than 2,000 chemicals that are registered annually by the Environmental Protection Agency (EPA) and used in commerce (we now have adequate human toxicity data on less than 40 percent of these). Much of this new data will not provide unequivocal answers but will require the development of new interpretive expertise and mechanisms to deal with problems such as false positives, which could lead to inaccurate diagnoses and intervention errors.
Finally, though the costs of the chips can be expected to drop, there may be a period of time when they would be available only to the wealthy. This period of time could be much longer if the health care system refused to underwrite their use, making early detection and associated intervention options unavailable to the uninsured and low-income portions of the population who might have high exposures to environmental toxins. This situation would also be found in less developed countries with dirty industries and poor environmental laws, where populations may have few options to monitor exposure and ultimately escape disease. Who will decide who benefits and who does not?
This is clearly a situation where rapid scientific and technological advance could outrun our institutional capabilities and test our moral fabric. As we all know, social innovation and moral development do not obey Moore's Law. The most important question is not whether such technologies will be developed and applied (they will) but whether we will be ready as a society to deal with the associated ethical, institutional, and legal implications. Steve Fodor of Affymetrix, one of the leading manufacturers of gene chips, recently remarked that, "Ninety-nine percent of the people don't have an inkling about how fast this revolution is coming." Although there has been a recent flurry of attempts by a wide variety of think tanks and policy analysts to "reinvent" the regulatory system, there is no indication that the environmental policy community is paying attention to this development.
This brings us to the final and most important lesson of the gene chip. It was only 35 years ago that Herman Kahn and his colleagues at the RAND Corporation confronted the policymaking community with the possibilities and probable outcomes of another of our large scientific and technological enterprises: The Manhattan Project. By outlining the potential outcomes of a war fought with thermonuclear weapons, they taught us two important things. First, science, and especially big science like the Human Genome Project, has far-reaching effects that are often unintended, unanticipated, and unaddressed by the people directly involved in the scientific enterprise. Second, and probably more important, is that better foresight is possible and can lead to better public policies and decisionmaking. Though the pace of technological change has accelerated, we have forgotten Kahn's lessons. The elimination of the Office of Technology Assessment in 1996 helped ensure that we will continue to drive through the rapidly changing technological landscape with the headlights off. In times like these, we need more foresight, not less. Embedded in the intriguing question of how the gene chip might affect environmental policy is the larger question of who will ultimately protect us from ourselves, our creations, and ultimately, our hubris. We are placing ourselves in a position described so well over 100 years ago by Ralph Waldo Emerson when he wrote that, "We learn about geology the day after the earthquake."
Wendy Yap, a fellow of the American Association for the Advancement of Science, has been working at EPA. David Rejeski of the EPA Office of Policy is serving as the executive director of the Environmental Technology Task Office at the Office of Environmental Quality.