Motivation for the proposals

Humans have been selectively breeding animals, and thereby intentionally altering their genomic DNA, since the dawn of domestication. Mutagenesis (chemical or radiation-based) followed by trait-based selection has been used for plants, and, to some extent, animals, for several decades.^{Reference Purdom, Lincoln and Schröder5} The FDA excluded both selective breeding and mutagenesis from its definition of intentionally altered genomic DNA.

Genetic engineering via splicing DNA sequences from one species into another (recombinant DNA techniques, “transgenes”) were first demonstrated in the early 1980s.⁶ More recently, gene-editing techniques such as CRISPR have been developed.^{Reference Wang7} These are designed to introduce edits at precise locations in the genome, compared to, in the FDA's words, “the more random changes associated with genetic engineering.”⁸ Of these modern technologies, CRISPR in particular is simple and cheap, making barriers to its use very low, with do-it-yourself CRISPR kits aimed at biohackers now available.

Genetically altered animals have been developed or proposed, among other purposes, for research to better understand human health and disease, to improve the food supply, to produce pharmaceuticals, to make pets, to bring back extinct species, to reduce the impact of invasive species, and to control populations that spread disease.

Already developed examples in the food category include the previously mentioned dairy cattle gene-edited to not have horns, salmon that grow much faster than normal due to inclusion of a gene that boosts growth hormone levels,^{Reference Entis9} pigs resistant to certain viruses,¹⁰ chickens genetically engineered to not transmit avian flu to other birds,^{Reference Lyall11} and low-fat pigs, produced through the inclusion of a gene that most other mammals have that allows them to better regulate their body temperature.^{Reference Stein12} In the pet category are Glo-Fish, aquarium fish that glow due to the inclusion of fluorescent protein from other species, and micro-pigs.^{Reference Cyranoski13} Animals developed for pharmaceuticals include a goat that produces an anticlotting protein in its milk, a chicken whose eggs contain a drug for a specific cholesterol disease, and pigs modified to bring the prospect of pig organs for human transplant one step closer.^{Reference Fischer14} Combating genetic disease due to in-breeding of purebred pets is a target for gene-editing. For example, almost all Dalmatians have a disease called hyperuricemia, which causes stones that can block the urethra and can lead to a burst bladder. The genetic edit to cure this disease is being pursued by a biohacker.^{Reference Rosenblum15} Other animals under development include mice modified to make them immune to Lyme disease,^{Reference Goldberg16} gene drives against invasive rodents,¹⁷ and the de-extinction of species such as the passenger pigeon and wooly mammoth.¹⁸ Of these examples, some are based on the older genetic engineering technology of recombinant DNA, and some are enabled via modern gene-editing techniques.

These examples demonstrate that the underlying genetic change can range from giving more animals healthy alleles that are already existing in that breed (the Dalmatians), to introducing alleles from closely related breeds (the hornless cattle), or from related species (the low-fat pigs). They include the removal of certain genetic elements (the pigs for possible organ donation). And they include the addition of genes from more distantly related species (the GloFish). The possibilities are vast.

Background to the Proposals

The history of regulating animal genetic modifications in the United States started in the 1980s in response to the first successful interspecies gene transfers, using recombinant DNA. These successes prompted questions of thow to regulate their myriad potential commercial applications. In response, the Reagan administration assigned oversight responsibility to three agencies — the FDA, the U.S. Department of Agriculture (USDA), and the Environmental Protection Agency (EPA). The administration set out its vision in the Coordinated Framework for Regulation of Biotechnology in 1986, subsequently updated in 1992 and again in 2017.¹⁹ These new regulatory oversight responsibilities were framed as extensions of the agencies' respective existing powers. The updated Coordinated Framework states that the agencies will only exercise regulatory authority over biotechnology products based on cost-benefit risk analysis, irrespective of the technology used to produce them.

In 2009 the FDA issued guidance #187 aimed at bringing all genetically engineered animals under its review, not only those for consumption or altered to produce pharmaceuticals.²⁰ The move was prompted by the commercialization of these animals being “no longer ‘over the horizon’.”^{Reference Wadman21} The 2009 guidance justified this expansion of jurisdiction through a novel interpretation of animal drugs in the Food, Drug, and Cosmetics Act of 1938.²² The definition of an animal drug includes “articles intended for use in the diagnosis, cure, mitigation, treatment, or prevention of disease in man or other animals;” and “articles (other than food) intended to affect the structure or any function of the body of man or other animals.” Being regulated as a drug involves product development standards, an extensive approval process, and post-approval responsibilities. This would apply to the originally developed animals and all their progeny for perpetuity. The FDA has to date approved just three animals: the goat (2009) and chicken (2015) for pharmaceutical applications, and the fast-growing salmon (2015). It deliberated over the salmon for 20 years, raising concerns about the process.

The regulation of genetically engineered animals can be contrasted with that of plants. Whereas the underlying technologies are the same, and the potential consequences on the environment and on human health may be similar, no semblance of this comparability appears in the regulatory framework. The USDA reviews genetically-engineered plant products under the Plant Protection Act to assess risk to plant health. Once a crop passes USDA review, it may undergo a voluntary review by the FDA. As of December 2017, 178 genetically altered crops had been approved through this process.²³ The USDA regulatory approach was based explicitly on recombinant DNA, and does not extend to gene-editing.²⁴ This process-based approach, rather than a risk-based approach, has enabled around 30 gene-edited plants to bypass regulatory oversight over the last five years.^{Reference Waltz25}

In light of the advent of gene-editing technologies, in July 2015 the Obama administration launched a process to modernize the 2009 guidance on biotechnology regulation. They commissioned a report from the National Academies of Science (NAS) on the future of biotechnology, focused on risks and how to assess them. They also called for the roles of the different agencies to be clarified, and for a long-term regulatory strategy to be developed.

On January 19, 2017, President Obama's last day in office, a flurry of guidance proposals were published. The FDA put forward three proposals: an updated version of guidance #187, extending guidance covering animals genetically engineered with recombinant DNA constructs (“transgenes”) to cover gene editing;²⁶ a proposal ceding regulation of mosquito population levels to the EPA; a proposal that the regulation of genetically edited plants would be no different to genetically engineered plants, which is on a voluntary basis.²⁷ Meanwhile the USDA proposed updating its regulation of genetically engineered plants. All proposals were opened for a period of public comment.

In November 2017, after reviewing the comments received from the public, the USDA announced that it was withdrawing the proposed guidance.²⁸ Instead, it announced it will begin again by re-engaging with stakeholders. The USDA disclosed that 203 comments were submitted on the draft guidance including both criticism that the proposed guidance would unduly lengthen the time to market of technologies, as well as that the review process was “insufficiently rigorous.”

The Backlash to the Proposals

The 151 comments the FDA received in response to its draft guidance concerning animals represents a snapshot of a broad range of interests and concerns, which we summarize here. The majority (117) of responses came from individuals, 14 from organizations associated with farming, 7 technology companies, and 13 others, including consumer organizations.²⁹ Fewer than ten were supportive of the FDA's approach. A minority (~20) were neutral. The vast majority (~80%) opposed the proposed guidance.

The most common argument opposing the guidance - which appeared independently in at least 50 responses - was that the product, and not the process that created it, should be the focus of regulation. Most stated further that such product-based regulation should be based on an assessment of risks versus benefits. These voices see the proposed guidance as a move away from the coordinated framework. Several commenters referenced the approach suggested in the NAS report. The report recommended a risk-based framework taking into account the complexity of the product and its familiarity compared to existing products whose risks are assumed to be better understood.³⁰ Several commenters expressed concern that the regulations as proposed will stifle needed innovation and ensure that only the largest bio-technology firms will be able to participate. One of the most damning comments came from the National Association of State Departments of Agriculture (NASDA), whose members are co-regulators in this space. It called the proposals a “reversal of policy” lacking risk-based justification and running counter to the administration's policy of reducing regulation and promoting innovation.

Another repeated concern was that the process-focused approach fuels anti-science fear mongering. A representative comment stated that the proposed guidance “demeans scientists, misinforms the public about biological sciences, and regulates based on fear and trembling rather than identified risks.”³¹

The Centre for Food Science, a public interest and environmental advocacy organization, submitted comments signed by 23,777 members (in a single submission), with a further 12 individuals, the Friends of the Earth (on behalf of their over 1 million members), and the Consumers Union submitting the same or similar comments. They claimed that the FDA “does not have authority under its enabling statute to regulate GE animals or insects absent a GE-specific statute passed by Congress,” and they called for a moratorium on the approval of all genetically modified animals prior to such legislation. They also emphasized understanding the risks posed by these new technologies. A small number of individuals voiced outright opposition to any genetic modification of animals.

A dozen individuals opposed the guidance arguing it severely hinders our ability to help cure genetic disease in animals, as in the Dalmatian example.

Three technology firms working on genetically modifying animals for non-food applications called for separate regulation for non-food products. The National Aquaculture Association and the California Department of Fish and Wildlife wrote with concerns that the guidelines as proposed would inappropriately cover established treatments used to induce desirable chromosomal abnormalities in fish.

Responses from all sides showed frustration at the lack of coordination across the different federal agencies. The Centre for Food Science called for a cohesive interagency framework and NASDA lamented “the apparent lack of coordination between FDA and its sister agencies in the Federal government.” The NAS report tactfully stated that the existing regulatory approach “could be considered to appear fragmented.”³²

The Proposals should be Withdrawn

We agree that the product should be the subject of regulation, and not the process that created it. A focus on product would force the FDA to clarify its regulatory intent, increasing transparency not just for product developers, but also for the public. It would elevate the importance of understanding risk, a view that was shared by a broad cross section of commenters. By contrast, a focus on process bunches diverse applications under one umbrella, from curing inherited diseases to adding in genes from distant species, from food to de-extinction of animals. And it divides products that should be considered together. In contrast, the USDA excluded from its proposed regulations any genetically modified organism that could have been produced using traditional breeding techniques because “[s]uch organisms are essentially identical, despite the method of creation.”³³

Moreover, history has shown that biotech regulations focusing on process have a frustratingly short shelf life. Regulation introduced in 2009 focused on process, calling out recombinant DNA. Just a few years later, new technologies arrived that are not covered by that regulation. Updating guidance with updated technologies repeats the original error. Already technology is outpacing proposals: mice have had their epigenomes successfully edited, and as these terms are currently understood they would not qualify as having “intentionally altered genomic DNA”.^{Reference Liu34}

A focus on process has allowed the FDA to avoid answering tough product questions. This needs to change. The approach proposed by the NAS in its report concerning the future products of biotechnology, in our opinion represents a sound core for a revised approach: a single cross-agency entry point for the risk based appraisal of new products. In the context of technology designed to affect animals, an additional concern should be for animal welfare. Technology such as gene editing has huge potential to not only improve animal welfare, but also to decrease it.

The lack of consistency across different agencies concerning genetically altered organisms is not only confusing within the U.S., it makes it almost impossible for other countries to follow the U.S.'s lead. The aim should be internal consistency within the U.S. as part of a cohesive international approach for these international issues: protection from large-scale risk while helping provide for the food security and better health of humanity.

Finally, the FDA received a similar number and spread of comments as the USDA received in reaction to its proposals. The same arguments cited by the USDA in withdrawing its own proposals apply equally to the FDA. In both cases, the proposals clearly fail to attract even minimal support from a broad range of commenters. Moreover, as the USDA noted, the publication of the proposed rules constrained its ability to explore alternatives with stakeholders. The withdrawal opened up the opportunity for “a more open and robust policy dialogue.”³⁵ Given the opposition to the proposals, it would be consistent for the FDA to likewise withdraw them as a first step towards a cohesive framework. Then we can start the much needed dialog to build a risk-based approach, both for gene-editing and for the biotechnologies to come.