Genetically Engineered Salmon – Coming Soon to a US Store Near You?
By Jessica Brandt and Ashton Merck
After almost 25 years of navigating rough regulatory seas, AquaBounty Technologies (ABT) announced its first sales of AquAdvantage salmon in Canada in August 2017. It is the first genetically engineered (GE) animal product intended for human consumption to reach the consumer market in North America and could reach US shelves soon.
To date, AquAdvantage salmon have faced a lengthy regulatory approval process, including significant opposition. While ABT argues that AquAdvantage is “the world’s most sustainable salmon,” and both US and Canadian authorities eventually deemed it acceptable for consumer health and the environment, some of its most vocal critics have been environmental advocates concerned about the resource demands of raising the fish and the carbon footprint of shipping it across countries, consumer groups who want labeling of all genetically engineered food products, and members of the commercial salmon fishing industry who seek to differentiate their product from its genetically engineered counterpart.
AquAdvantage salmon have been genetically manipulated to allegedly grow at a quicker rate than wild-type Atlantic salmon, via the insertion of a recombinant gene construct. This method is a process commonly used in laboratory settings to study the importance of a given gene or set of genes. For the AquAdvantage salmon, the construct contains genes from two other species of fish: the growth hormone gene from Chinook salmon, and the promoter gene from ocean pout.
In this post, we provide an overview of these arguments in the context of US regulation, paying particular attention to the issues of environmental impacts and food safety. We conclude with a brief summary of the Canadian system and process of approval; however, more extensive comparative research is needed in order to understand the extent of variations in the regulatory process and political environment in the U.S. and Canada. It is also worth noting that this comparison does not address the regulatory environment in Panama, the location where the AquAdvantage fish are raised in land-based pens.
How is AquAdvantage regulated in the US?
Because AquAdvantage is designated for human consumption in the US, it is regulated by the US Food and Drug Administration (FDA). Products like AquAdvantage pose complex challenges for the FDA, as the agency operates under a broad mandate dating back to the early twentieth century that may not always be a perfect fit for new technologies. The growth promoters in AquAdvantage fit the FDA definition of a “new animal drug,” because the genetic changes affect the animals’ physiology, and the genetic modification had not been used for an extended period of time (in contrast to FDA’s regulations for an “animal drug”). Critics have petitioned the FDA to regulate the genetic modification as an (unsafe) food additive. However, the statutory definition of “food additive” specifically excludes new animal drugs. A report produced by the Congressional Research Service suggested that regulation as a food additive might require more comprehensive toxicological studies than currently mandated through the NAD approval process, but what is meant by that statement is not clear. In a response to public comments requesting the FDA to regulate GE salmon as a food additive, the FDA noted that there is “no substantive difference” in the food safety standards of approval as a food additive from that of a new animal drug. Both rely on the statutory definition of “safe” to mean “reasonable certainty of no harm” from the Federal Food, Drugs, and Cosmetic Act (FFDCA).
Since 1995, when ABT submitted the initial application for a New Animal Drug to the FDA, the company has made two key claims about the genetic modification: first, that the combination of these two genes caused fish to grow twice as quickly as wild-type Atlantic salmon, and second, that these fish have a more efficient feed conversion ratio than wild salmon, with each pound of feed adding a pound of flesh (representing a 20-25 percent higher feed conversion rate than other farmed fish).
The FDA spent 20 years evaluating these claims: first, based on its statutory mandate to protect consumers from “adulterated” and “misbranded” foods under the definition of a “new animal drug,” under the US Federal Food, Drug, and Cosmetic Act, and through an environmental assessment under the US National Environmental Policy Act (NEPA). To conduct their investigation, the FDA inspected the hatchery and grow-out facilities, convened a Veterinary Medicine Advisory Committee, consulted with National Oceanic and Atmospheric Administration’s (NOAA) National Marine Fisheries Service and the US Fish and Wildlife Service, completed an environmental assessment under NEPA, issued a proposed decision, and hosted a public-comment period on the proposal that received more than 320,000 replies. After gathering and analyzing all of this evidence, the FDA determined that AquAdvantage salmon are as safe and nutritious as conventional salmon, and granted final approval on November 19, 2015.
What did the US FDA approve?
The parameters outlined in the new animal drug approval allow initial production only in a limited geographic scope. Contrary to concerns that FDA approval might give ABT too much latitude, the FDA effectively gave AquaBounty a one-time approval for specific genetic modifications, in a specific setting, for the specific firm. The appendix of the FDA approval explicitly prohibits AquaBounty from raising their salmon in the United States, and from expanding their facilities, without submitting another application. Moreover, another company cannot license the technology to grow their own salmon without getting their own, separate approval from FDA.
Currently, the FDA approval does not allow for the AquAdvantage salmon to be raised in the US. Instead, AquAdvantage eggs are fertilized at the broodstock facility on Prince Edward Island, Canada, and shipped to the Panamanian highlands where the fish are raised in land-based recirculating aquaculture systems, harvested from these land-based tanks once they reach market weight, and shipped to distributors for retail sale. In the future, if ABT or another firm requests to raise such salmon in the United States (or in open water rather than in land-based tanks), the company would need to submit another New Animal Drug application, and the FDA could decide to conduct additional analyses of health and environmental effects. The FDA approval also stipulates precise measures to prevent the fish from escaping, including through reporting and monitoring by AquaBounty personnel.
Initially, the FDA released no specific guidance on labeling the fish for sales to consumers, based on the finding that the AquAdvantage salmon would be as safe and nutritious as conventional Atlantic salmon. The agency subsequently released draft guidance for voluntary labeling, which appears to have been preempted by a statute that delegates responsibility of labeling of all genetically engineered food products to the Agricultural Marketing Service in the USDA. This statute, and the overall context for the debate over labeling of genetically modified foods, is discussed in more detail below.
In 2002, the National Academy of Sciences’ National Research Council issued a report on the science of animal biotechnology. At the time, the FDA was seeking guidance on how to regulate cloned cattle, among other genetically engineered products. The NAS report concluded that the environmental concerns, particularly the possibility of escape by genetically engineered fish and other “mobile” animals, were of a higher priority than food safety concerns.
When the FDA conducted an Environmental Assessment (EA) of AquAdvantage salmon under NEPA, it limited its assessment to the environmental impacts inside the US. Under longstanding US legal doctrine, federal laws are presumed not to apply in other countries unless Congress has expressly required such extraterritorial reach in the language of the statute. This doctrine is based in part on avoiding conflict with the domestic laws of other sovereign countries. Because AquAdvantage salmon are entirely produced outside of the US, the FDA studied impacts in the US and found that “[o]ur analysis in this EA has determined that the agency’s action will not significantly affect the physical environment of the United States.” The FDA focused on the risks posed to wild populations (via gene transfer or ecological competition) that could arise should ABT’s fish escape, and much of the US approval sets strict requirements for monitoring and control of fish populations. To diminish these risks, the company only raises female fish containing three chromosome copies (i.e. triploid fish), thereby rendering them sterile, in land-based pens. However, some scientists argue that the biological containment approach employed by AquaBounty is not always effective.
Though outside the scope of the FDA’s US-focused mandate, there are additional environmental considerations with this product. For example, there can be heavy freshwater and electricity demands when re-creating ocean conditions in land-based aquaculture facilities, and though the company has mitigation efforts in place (for example, tank covers, nets, filters, and screens), the incineration of fish wastes and release of effluents to local watersheds are sources of environmental pollution. (These concerns could apply to all aquaculture fish production, not only to the AquAdvantage fish or to genetically modified fish.) In general, aquaculture has relied heavily on fish protein and oil inputs. Historically, there have been concerns about the burden these inputs to aquaculture placed on wild fish populations, especially open-ocean small-bodied pelagic fish (e.g., anchovies, sardines, mackerel). More recently, however, these concerns have diminished as aquaculture has begun to more sustainably source these inputs, including the recovery and recycling of fish wastes and processing by-products.
Meanwhile, the FDA’s narrow scope of assessment also limited consideration of the product’s potential benefits. As other scholars have argued, the larger supply of salmon enabled by genetic innovations such as AquAdvantage appear likely to displace other proteins (notably beef) in consumer markets, thereby reducing adverse environmental impacts from raising livestock –including those related to feed inputs and methane emissions. Substituting fish for red meat-derived protein in the diet could also carry health benefits, such as reductions in coronary heart disease risk, which are tied to the high levels of omega-3 fatty acids in the diet. AquaAdvantage salmon may also reduce the risk of contaminants in fish, including mercury. By raising the GE fish in land-based aquaculture facilities that minimize atmospheric contaminant deposition (among other types of contaminantion), the cardiovascular health benefits of GE salmon may be quite large. Finally, fish also present efficiency advantages over land-based animals with 64-68 percent of the average salmon being consumable compared to only 40 percent of a cow, 52-56 percent of a pig, and 46-58 percent of a chicken.
The FDA’s limited assessment, focused on replacing non-GE salmon with GE salmon and the impacts in the US, did not encompass these broader effects on total production and consumption of salmon and its substitutes such as beef. Moreover, whereas AquAdvantage salmon may travel long distances before reaching US stores, its carbon footprint is predicted to be relatively small when compared to Chilean and Norwegian farm-raised salmon consumed in the US, and especially when compared to animal-based protein alternatives including pork and beef. 
Additional health effects on consumers
A key question for FDA was whether AquAdvantage salmon posed health risks or benefits different from conventional salmon. Under the FFDCA, an NAD can be approved as “safe” if FDA finds there to be “reasonable certainty of no harm.” This is the same definitional criteria of safety that is used for other foods and drugs regulated by the FDA, including food additives. In order to evaluate potential variations in health risks relative to Atlantic salmon, the FDA first needed to identify whether or not AquAdvantage salmon fit the regulatory definition of Atlantic salmon. Using the Regulatory Fish Encyclopedia, which includes images, physiological descriptions, and other taxonomic indicators useful for regulators to identify potential fraud or deception, the FDA determined that AquAdvantage salmon fit the definition of an Atlantic salmon.
One of the concerns about GE foods is that the genetic modification may pose new or unexpected allergenicity risks, which is important in this case considering that fish allergies are among the eight most common allergies. The FDA conducted its own analysis of diploid and triploid AquAdavantage salmon in an analysis of six fish per group, and found that four diploid (66.7%) and one triploid (16.7%) had a mean allergenicity potential above the threshold level. From this evaluation, FDA concluded that the triploid fish (those intended for human consumption) did not pose a greater allergy risk than the diploid fish. However, this conclusion was not based on comparison to the control group, and the control salmon were not farm-raised fish. In addition, these results were based on a very small sample size, and they do seem to indicate at least slightly increased allergy risk. Despite recognizing that “[t]here are insufficient data and information to draw a conclusion on the relative allergenic potency of diploid ABT salmon,” the FDA did not request additional product testing. Instead, they recommended that people allergic to conventional salmon should also avoid the AquAdvantage salmon.
Where is AquaBounty now?
The FDA approved AquAdvantage salmon for human consumption in 2015, but its entry to the US market has been delayed by subsequent demands for improved labeling and notification measures from both consumer advocacy and representatives of the commercial salmon fishing industry. Despite FDA’s finding that AquaBounty salmon did not need to be labeled differently from ordinary Atlantic salmon, the agency released voluntary guidance in draft form to producers who may wish to label their product. Agency guidance is very similar to the concept of “best practices” and does not have the same force as a law or regulation.
However, the broader debate over the labeling of GE foods led to a Congressional action that effectively superseded the FDA guidance on AquAdvantage salmon. In response to the passage of a state-level labeling law in Vermont, a coalition of agricultural producers, retailers, and farm advocacy groups pushed Congress to enact a federal standard for labeling of GE foods. In 2016, Congress amended the Agricultural Marketing Act of 1946 to include labeling of genetically modified foods. This law, formally known as the National Bioengineered Food Disclosure Standard, shifted the labeling responsibility of all genetically engineered food products – not just AquAdvantage – from the FDA to the Agricultural Marketing Service of the US Department of Agriculture (USDA). This law offers three options for labeling GE foods: a QR code, a symbol (such as that for the USDA Organic label) or a statement. Critics of the new law, concerned that industries and retailers will overwhelmingly favor the QR code over text labeling, have colloquially referred to this law as the “DARK,” or “Denying Americans the Right to Know,” Act. Since the signing of the National Bioengineered Food Disclosure Standard, many producers have begun complying with one of the three options outlined in the law, even as USDA moves forward with the rulemaking process.
While consumer concerns about GE foods, such as AquaBounty salmon, did not prevent the federal regulatory approval of AquAdvantage in the US or in Canada, the labeling rules may keep the product off U.S. grocery shelves for the near future – in part at the behest of the incumbent salmon industry in places like Alaska, seeking to fend off competition. Meanwhile, numerous companies have publicly announced their decision not to stock or carry GE seafood, largely at the behest of advocacy group Friends of the Earth. Among them are major retailers such as Costco, Whole Foods, Trader Joe’s, Kroger, Safeway, and Target.
How was Canada’s regulatory process different?
Upon initial examination, there does not appear to be an obvious difference in either the regulatory or political environment in the US and in Canada that would easily explain why AquAdvantage was sold to Canadian consumers before it reached consumers in the United States. In fact, the press release announcing Canadian approval of AquAdvantage salmon on May 19, 2016, takes care to note that Canada “is not the first country to approve this product,” as the US FDA had approved AquAdvantage salmon for human consumption several months earlier, in November 2015.
The Canadian regulatory environment is generally comparable to that of the US, in that the authority of approval for a product like AquAdvantage salmon was divided among multiple agencies with different types of scientific and technical expertise. Environment Canada reviewed the environmental impacts of AquAdvantage salmon, Health Canada reviewed the potential risks to human health, and the Canadian Food Inspection Agency reviewed the salmon’s safety for use as livestock feed. Health Canada evaluated the safety of AquAdvantage based on guidelines from the Codex Alimentarius, which is a set of globally recognized food safety standards promulgated by the Codex Alimentarius Commission within the Food and Agriculture Organization. In cooperation with Environment Canada and Health Canada, Fisheries and Oceans Canada also completed a risk assessment for AquAdvantage salmon which was peer-reviewed by an independent scientific panel.
Several consumer and environmental groups in Canada registered their opposition to the entry of AquAdvantage salmon by challenging the approval in court; they have thus far been unsuccessful. A private member’s bill was introduced in the Canadian House of Commons which would have amended the Canadian Pure Food and Drugs Law to require labeling of genetically modified food. This bill was decisively voted down on May 17, 2016.
The big picture
The ABT salmon case illustrates the challenges that regulators face when confronted with new products that do not easily fit into preexisting rules or regulatory regimes. FDA’s response has been to define its role narrowly – evaluating the health and environmental impacts only with regard to new salmon compared to conventional salmon, and only in the US. A broader analysis would encompass the larger portfolio of health, safety, and environmental impacts, both pro and con. Meanwhile, the drawn-out debate over labeling and its subsequent assignment to the USDA Agricultural Marketing Service through the legislative process illustrates how new technologies can face new demands for information and regulatory action, both by consumers and by industry incumbents.
The history of AquAdvantage salmon also provides a cautionary tale about the importance of balancing the development of innovative solutions with attention to consumer values and expectations. More generally, the broader negative response to genetic engineering in food speaks to an ongoing need to build public trust around scientific achievements through a culture of transparency and communication. However, the relatively brisk sales numbers in Canada suggests that consumer opposition to AquaBounty may be less widespread than popular protests seemed to imply. As of August 4, 2017, Canadians have eaten 4.5 tons of AquAdvantage salmon.
More research on the risk-risk tradeoffs (both co-benefits and countervailing harms) presented by products like AquAdvantage is needed in order to fully evaluate the risks and benefits. This is especially important in the food and agricultural sector, where some predictions say that increasing population sizes will require us to double our food supply. Demographic shifts may result in greater demand for animal protein-intensive diets, even as efforts to mitigate anthropogenic climate change will require the development of less carbon intensive and more sustainable options.
Jessica Brandt is a 5th year Ph.D. Candidate in the Environmental Health Program at Duke University. Her research focuses on the impacts of coal ash waste streams in North Carolina lakes.
Ashton Merck is a Ph.D. Candidate in the History Department at Duke University whose dissertation analyzes the evolving scientific, technical, and legal challenges in food safety and international trade over the past half-century.
The authors wish to thank Profs. Jennifer Kuzma, Martin D. Smith, and Jonathan B. Wiener for their feedback on earlier drafts of this post.
 See generally, James Harvey Young, Pure Food: Securing the Federal Food and Drugs Act of 1906 (Princeton University Press, 2014); I. Glenn Cohen and Holly Fernandez Lynch, eds., FDA in the Twenty-First Century: The Challenges of Regulating Drugs and New Technologies (Columbia University Press, 2015).
 21 U.S.C. 201(s)).
 Harold F. Upton and Tadlock Cowan, “Genetically Engineered Salmon,” Congressional Research Service, 2014. 1-31.
 21 U.S.C. 201(s)).
 Rosamond L. Naylor, Rebecca J. Goldburg, Jurgenne H. Primavera, Nils Kautsky, Malcolm CM Beveridge, Jason Clay, Carl Folke, Jane Lubchenco, Harold Mooney, and Max Troell, “Effect of aquaculture on world fish supplies,” Nature 405, no. 6790 (2000): 1017-1024.
 Rosamond L. Naylor, Ronald W. Hardy, Dominique P. Bureau, Alice Chiu, Matthew Elliott, Anthony P. Farrell, Ian Forster et al. “Feeding aquaculture in an era of finite resources.” Proceedings of the National Academy of Sciences 106, no. 36 (2009): 15103-15110; C. J. Shepherd and A. J. Jackson. “Global fishmeal and fish‐oil supply: inputs, outputs and marketsa.” Journal of fish biology 83, no. 4 (2013): 1046-1066.
 Martin D. Smith, Frank Asche, Atle G. Guttormsen, and Jonathan B. Wiener. “Genetically modified salmon and full impact assessment.” Science 330, no. 6007 (2010): 1052-1053.
 William S. Harris, Michael Miller, Ann P. Tighe, Michael H. Davidson, and Ernst J. Schaefer. “Omega-3 fatty acids and coronary heart disease risk: clinical and mechanistic perspectives.” Atherosclerosis 197, no. 1 (2008): 12-24.
 21 USC § 346a(c)(2)(A)(ii).
 This phenomenon can be described as “raising rival’s costs,” in which an incumbent industry seeks to raise barriers to entry. See Steven C. Salop, and David T. Scheffman, “Raising rivals’ costs,” The American Economic Review 73, no. 2 (1983): 267-271. For a broader review essay of mechanism choice in regulatory governance, see Jonathan B. Wiener and Barak Richman, “Mechanism choice,” in Research handbook on public choice and public law, edited by Daniel A. Farber and Anne O’Connell (Edward Elgar, 2010). 363-396
 A “private member’s bill” is a motion presented to the House of Commons or its equivalent brought by someone who is not a cabinet minister. For a variety of reasons, including that debate on these bills is time-limited, very few of these bills ever become law. See Nora S. Lever, “What’s Happened Under the New Rules?” Canadian Parliamentary Review, Autumn 1988, 14-16.