The legal and regulatory context for human gene editing.
The ecology of this system is one in which there are many legal or policy issues that combine to affect whether biotechnology is promoted or hindered in any particular country. It ranges from topics such as intellectual property rights, which are reflected in areas from patent policy, to international trade laws, which will have a huge effect on whether or not the new products are going to be able to cross borders easily and under what conditions. The regulatory framework is going to determine the speed at which biotechnology moves from laboratory to development to marketed product.
The consumer demand will also be a profoundly important feature in determining which products are developed, because so many discoveries do not lead to something that the public wants or needs, or that it knows it wants and needs. This will also be affected by variables such as stigma and cultural attitudes.
Last of course, but certainly not least, are areas of public research and investment. All of these together are going to combine into a vision of how a particular country moves or does not move biotechnology. Some of the categories that have been proposed by other scholars range from promotional, in which a country is actually pushing the innovation; to a more neutral stance, in which it simply proceeds or not with as little government direction as possible; to precautionary; to an absolutely prohibitive system that either defunds entirely or even makes criminal the technology.
It is worth keeping in mind that within a country, one can have very different attitudes about different aspects of biotechnology. For example, the United States has a fairly permissive approach to biotechnology applied to genetically engineered animals and plants in the agricultural sector, whereas it has a much more cautious approach when it comes to the use of biotechnology in the context of human clinical care and therapies. There does not have to be a single approach to biotechnology across all application areas. There can be differences among countries and even within a country.
One can also look at how different areas of policy can be tied to one or another of these visions of an overall biotechnology direction. For example, strong patent protection can be viewed as promotional because it gives industry the greatest possible financial incentive to pursue particular application areas. However, from the basic science and research community point of view, strong patent protection can sometimes be perceived as slowing the ability to collaborate or take advantage of one another's work.
In the area of biosafety, we see more case-by-case evaluation of biotechnology products, where everything really begins to hinge simply on the presumption about risk. One can take a precautionary approach that presumes it is dangerous until it is proven safe, or a permissive approach that presumes it is safe until it is proven dangerous. Since it is often impossible to prove either danger or safety, where that presumption falls will often be more determinative than anything else in deciding how quickly technologies move from the basic science laboratory to clinical research to application.
Finally, in the area of public information, there is a very lively debate going on, particularly in the United States, about the labeling of foods that have some component that involves modern biotechnology. For example, now that the Food and Drug Administration (FDA) has approved the sale of a genetically modified farmed salmon, there is a debate about whether that salmon has to be identified for consumers.
If we have systems that carefully distinguish between those things that are the products of modern biotechnology and those that aren't, we could be setting ourselves up for a more precautionary regulatory approach because it will tie into public attitudes that are often based on concern about either the corporate influence or the actual underlying science. On the other hand, if regulation is mandated only when there is evidence of a higher level of risk, products will reach the market more quickly, reflecting a more promotional stance.
To implement any one of these approaches, we have a variety of mechanisms that range from the least to the most enforceable. Public consultation is the least enforceable approach, and there is a spectrum of regulatory and legislative measures that can strengthen the level of control.
In the area of public consultation, we have numerous examples from around the world. In the United States, the National Environmental Policy Act is unusual among environmental laws because rather than telling individuals or companies what they can and cannot do, it simply provides that when the government makes a particular decision, it must be subjected to a higher degree of public scrutiny than is typical. The catchword for this approach is that "sunlight is the best disinfectant." By incorporating public comment, it creates political pressure that can drive decisions in one way or another, and it allows for some interplay between government expertise and public consultation. We see other examples of it in the approval process for products such as engineered salmon, which required a number of public hearings.
Canada, when it looked at assisted reproduction, formed a royal commission on new reproductive technologies that held hearings on the topic across the country. In the European Union (EU), genetically engineered foods, or GMOs as they are usually referred to there, are of special concern. There is actually an EU directive requiring that there be a degree of public access to information whenever a product potentially affects biodiversity or other environmental elements.
Public consultation is considered an alternative to a centralized directive form of governance. One simply creates the situation in which the public can, through its own decentralized processes, exert pressure on government or on industry and thereby alter the direction or the speed of biotechnology innovation.
Next in this hierarchy of enforceability comes voluntary self-regulation. The 1975 Asilomar conference on recombinant DNA technology was one of the more notable examples of voluntary self-regulation by the scientific community when it recognized that there were certain risks that needed to be investigated before it pushed forward at full speed.
The research community voluntarily imposed on itself moratoria on certain applications and implemented a series of precautionary measures having to do with containment of possibly dangerous materials. A more recent example is the set of guidelines for human embryonic stem cell research, which were developed by the U.S. National Academies and the International Society for Stem Cell Research.
What is interesting about these instances of self-regulation is that unlike the government-imposed rules, these were truly self-imposed rules that were seriously constraining in many ways. They often called for prohibiting payment for certain materials and services in ways that limited the ability of the scientific community to move as quickly as it might want. For example, it limited the use of chimeras and established strict guidelines on the distribution of the gametes and embryos needed for research.
It was a success in the sense that it forestalled what might have been really onerous government action at the state or federal levels, and it demonstrated that self-regulation could be flexible and nuanced without sacrificing reliability. The self-regulatory approach has also been used in the case of "gain of function research," a very awkward name for research that increases the pathogenicity, transmissibility, or resistance to countermeasures of known pathogens.
Interestingly, these kinds of voluntary self-regulatory activities often lead directly into some government adoption by proxy of much of the content of the self-imposed rules. For example, in the gain of function area, some of the self-imposed rules led to a National Academies report, which then led, in turn, to the creation of the National Scientific Advisory Board for Biosecurity, which collaborates with its counterparts around the world to manage situations where there is fear that publishing key data will facilitate the transformation of useful biotechnology into bioterrorism.
There are government guidelines in other areas as well. These provisions technically are not enforceable, and yet they are very strongly persuasive because complying with them creates what essentially is a safe haven for companies. They know that if they stay within these guidelines, they are not going to run afoul of some actual regulation or law. These guidelines also create strong social norms.
At the international level, there is the Council for International Organizations of Medical Sciences (CIOMS), which is very influential in creating global standards for research on human subjects. It refers back specifically to the Nuremberg protocols and has the ability to be more restrictive than any particular national set of rules.
That doesn't mean that national laws will necessarily follow, but it establishes a norm from which nations feel free to deviate only when they can provide justification that it is necessary to achieve some public benefit. Therefore, the CIOMS becomes extremely influential, even if not enforceable.
At the far end of the spectrum, of course, we have regulation and legislation. For example, many nations have laws that specifically ban human cloning, although the United States is not one of them. That is not to say that it actually happens in the United States; it is just that there is no U.S. legislation that explicitly bans it. The U.S. regulatory system could, in theory, approve it, but it has never indicated any particular willingness to do so. Effectively, it is impossible to do it legally in the United States, but it is not considered a ban.
We should keep in mind that legislation has the advantage of being more politically credible, particularly in more or less functioning democracies, because it is seen as a product of elected representatives. On the other hand, legislation is extremely rigid and difficult to change. Once it is in place, it can be impossible to remove it, and it is often resistant to nuance. Therefore, it can be a very blunt instrument.
Regulation--that is, the detailed administrative rules adopted pursuant to legislative direction and authority--has the ability to be much more responsive and detailed, and is influenced to a greater extent by expert information. Yet, it also begins to become somewhat more divorced from public sentiment and begins to move into the world of the administrative state where there is rule by expert, which has its own challenges for democratic systems.
Focus on human gene therapy
Looking specifically at regulation of human germline modification, a 2014 survey of 39 countries by Motoko Araki and Tetsuya Ishii found a variety of regulatory approaches. Many European countries legally prohibit any intervention in the germline. Other countries have advisory guidelines. The United States has a complicated regulatory scheme that would make it very difficult to perform any germline modification. There are also funding restrictions on embryo research that might have a very strong effect on the underlying basic science needed to even get to the point of regulatory approval. And many countries have simply not considered the possibility.
There are international instruments that have been written at various levels to address aspects of genetics. For example, the Council of Europe's Oviedo Convention says that predictive genetic tests should be used only for medical purposes. It specifically calls for a prohibition on the use of genetic engineering of the germline or changing the makeup of later generations. It builds on earlier European conventions.
But like many international instruments, it is not ratified by every member country and, even when ratified, has not necessarily been implemented with concrete legislation. It has great normative value and can occasionally have enforcement-level value, but it is often lacking in the latter.
In the United States, gene therapy is handled in a regulatory system that treats it as a biological drug or a device, depending on its mode of operation. It comes under the comprehensive regulation of the FDA and under multiple laws focusing on infection control, efficacy, and safety.
The United States also seeks guidance from advisory bodies such as the Recombinant DNA Advisory Committee and the local research subjects review bodies that help to make sure that human clinical trials are managed in a way that agrees with the country's norms and regulations.
But what is perhaps distinctive about the United States is that although it has very strong controls in the pre-market stage of these technologies, once a drug, device, or biologic is on the market, the control becomes much weaker. That is, the United States regulates the products, but not the physicians who actually use those products. Physicians have the discretion to take a product that was approved for one purpose and use it for a different purpose, population, or dosage. There are some post-market mechanisms to track the quality of this work and to dial it back, but they are not as strong as in other countries.
Gene therapy in South Korea has a pathway very similar to the one in the United States. Interestingly, South Korea has come to have a focus on innovation, with expanded access to investigational drugs. It is also developing a system of conditional approval, which would allow for some use of a product prior to the accumulation of the level of evidence that is required in systems such as that in the United States.
Again, there are different versions of this. Even in the United States, regulators sometimes accept evidence from surrogate markers of effectiveness, which allows for a faster path to the market. Many other countries are also considering adopting some form of conditional approval.
The United Kingdom's (U.K.) system is a little different because not only is it operating within the context of the EU and its directives, but it has its own very strong pre-market review process. In addition, it has very strong post-market regulation of any procedures involving embryos or human fertilization. Thus, U.K. regulations cover not just the product, but also where the product can be used and by whom.
The EU has also added special provisions for advanced therapy medicinal products. Gene therapy is almost certainly going to be among them, so that there is an extra layer of EU review for quality control at a centralized level.
Japan has a regulatory pathway that tries to identify prospectively those things that are going to be high, medium, or low risk, and to regulate them accordingly. The United States follows a similar process in its regulation of medical devices.
But for drug regulation, the United States treats everything from the beginning as equally dangerous and runs every proposed drug through the same paces of testing for safety and efficacy. By contrast, in Japan, one will see an initial determination about the level of risk that is likely to be present for each proposed drug and the degree of stringency that the regulatory process must apply as a result.
Japan also has recently added a conditional approval pathway specifically for regenerative medicine and gene therapy products. It will be very interesting to see how this operates. It is still new, so the experience is limited.
There is certainly some concern that if new products are put into use too early in controversial fields such as embryonic stem cell research or gene therapy, a single high-profile failure might set back the entire field. In the United States, the death in 1999 of Jesse Gelsinger in a gene therapy trial at the University of Pennsylvania set back the field by years.
One of the challenges with the conditional therapy pathway is to balance the desire to move forward as quickly as possible while avoiding the kinds of adverse outcomes that not only injure individuals, but could slow progress to the point that many individuals who could have benefited in the future are denied the technology because it is delayed so significantly.
Singapore has a risk-based approach similar to Japans. What is interesting in Singapore is that it actually tries to figure out what would be a high-versus low-risk intervention in the area of cell therapy. The variables that are used include whether the manipulation is substantial or minimal, whether the intended use is homologous or non-homologous, and whether it will be combined with a drug, a device, or another biologic.
The only consideration one might add is autologous versus non-autologous use. In Singapore, these distinctions are used to classify the level of risk. In the United States, it is used to determine if the FDA has the jurisdiction to regulate that particular product.
Finally, Brazil provides an example of regulation and governance by accretion. It recently approved laws related specifically to genetically engineered foods, stem cell research, and cell therapy, but they are layered on top of earlier, more general rules: constitutional prohibitions on the sale of any kind of human tissue and 1996 laws on the patenting of human biological materials. Together they are creating a situation of confusion. The result is paralysis while people try to figure out how the laws are going to interact. It is a cautionary tale about how to proceed with legislation against the backdrop of older decisions that may have been made against different imaginary scenarios.
Product or process?
There is a fundamental divide in the world about how we regulate biotechnology that goes beyond the categories of promotional, permissive, or prohibitive. It is whether we think of biotechnology as a thing unto itself, or whether we think of it simply as one more tool that goes into making various products.
If one regulates the technology, one regulates everything about the technology in a comprehensive way. An example is the EU's community strategy, which takes a global approach to the technology that makes it easier for the public to understand the so-called "laws on biotechnology." One can focus on key aspects of the science that create key questions about the effects of a particular kind of innovation. Itto makes it possible to have consistent and overarching approaches to questions of great philosophical significance, such as what we mean when we say "human dignity" or "genetic heritage of mankind."
It also has the problem of needing much more specific legislation to focus on individual products because, as is noted in a contrasting system where you regulate the product and not the technology, as is the case in the United States, the technology itself is neither inherently dangerous nor safe. It is dangerous in some contexts and safe in others. In some products, it is easier to predict its effects. In other products, it is much less likely. Some products may have environmental impacts, and for others the impact will be confined to a single individual or a single animal.
Regulating by product gives one the advantage of being able to be much more specific about the degree of risk that is feared or anticipated, and the degree of caution needed, as well as being able to take advantage of mature degrees of expertise in the regulatory pathways appropriate for drugs, foods, and pesticides, and of the expert people who have been implementing those pathways for years.
The trouble is that it can be confusing to the public. If someone asks: what is the "law on biotechnology," the answer is that there are 19 different laws that cover drugs, devices, agricultural products, livestock, and so on. To many people, this sounds as if the country is not regulating biotechnology, and it creates the possibility for unintended or even unnoticed gaps among these laws or conflicts among them.
Whenever we are talking about this, whether in the human or non-human application, but particularly in the human, it is important to think about where in the R&D process we want to exercise control. Pre-market control is truly important to avoid the devastating adverse events that can occur if we move too quickly. But if pre-market control is too strong, not only does it slow the technology, but at a business level it creates a barrier to market entry for smaller players. Mature companies with large staffs know how to maneuver the regulatory system. A small company with very low levels of capital and a high burn rate is not necessarily going to be able to survive long enough to deal with a long and difficult pre-market process.
The AquAdvantage salmon that I mentioned earlier is made by a company that has reportedly been on the verge of bankruptcy during the 20-some years that the product was undergoing review. Another company in Canada that was trying to produce a pig that would be less environmentally damaging wound up abandoning this project, in part because that pathway was so long, slow, and expensive. There is a cost to pre-market controls that are so strong that they drive out the small, and often very creative, innovators.
One thing we have learned is that conditions on research grants, whether from government or philanthropies, can also serve as a strong regulator, but one that is much more responsive and much easier to adapt quickly to changing circumstances and changing levels of knowledge.
Finally, harmonization across national borders is crucial. If we want scientists to be able to use one another's materials, they have to have confidence that the materials were derived and managed in a way that meets everybody's common expectations of both ethical and biomedically safe levels of care.
We want to have uniformly high standards for research and therapy. We want to be able to reduce conflicts and redundancies in review procedures if we want the science to proceed in a way that is efficient as well as responsible. We learned this lesson with the many conflicts among jurisdictions in the area of embryonic stem cell research.
The more that we have effective systems for responsible oversight in the development and deployment of a technology, the more we can take chances. We can move a technology quickly because we have a chance to back up at the end and change course.
Innovation is not something that is in conflict with precaution. They are complementary strategies in which precaution will facilitate innovation and give us the confidence we need to support these new and risk-taking technologies.
R. Alta Charo is Warren R Knowles Professor of Law and Bioethics at the University of Wisconsin.