Policy commentary

Superior Salmon and More

Sep 27, 2010 | Randall Lutter

The first application to sell genetically engineered (GE) fish to U.S. consumers is nearing final consideration and its approval by the Food and Drug Administration would likely promote development and marketing of many innovative GE fish and meat products. How is the FDA dealing with the application and with public interest in labeling of such products?

Products from the biotech industry took a giant leap toward the market and the dinner plate earlier this week at two U.S. Food and Drug Administration (FDA) meetings on genetically engineered (GE) salmon intended for human consumption. The salmon, known as AquAdvantage and developed by AquaBounty, differs from Atlantic salmon currently raised in farms by having genes from a Chinook salmon and from an eelpout. This innovation enables it to reach market size in half the time as regular Atlantic salmon, a difference that may lower production costs enough to make a new class of onshore production facilities economical.

Such new production is potentially very big news, especially in light of the conclusions of FDA staff scientists that the product is as safe to eat as conventional farm-bred salmon. New controlled production facilities for new domesticated fish species may address the large and growing conflict between rising worldwide demand for fish and overharvesting of marine fisheries resulting from poor and ineffective regulation. Such facilities pose fewer environmental risks than the conventional fish farming practice of using nets or pens in relatively shallow marine waters and would render acceptable the environmental risks posed by AquAdvantage, which AquaBounty proposes to produce only as a sterile female fish. Given the cardiovascular benefits from fish, which is high in omega-3 fatty acids, greater fish supply (to the extent that it lowers the market price and increases consumption) should have indirect health benefits.

Approval and use of the AquAdvantage salmon also would pioneer a path, in principle, for other GE animal products to address vexing health and environmental problems. No country has yet allowed any product from GE animals to be sold for human consumption. But a number of innovative products intended for human food consumption are likely now under review at FDA including:

  • an “enviro-pig,” engineered to have exceptionally low phosphorus in its excrement, so as to address the problem of controlling phosphorous run-off from pig production facilities;
  • a “tuna-pig” high in omega three fatty acids; and
  • a cow resistant to bovine spongiform encephalopathy (mad cow disease).

Genetic engineering of farm animals, if regulated in a manner that adequately protects animal health and food safety and preserves incentives for research and development, offers the promise of low-cost solutions to a number of social and economic problems.

Whether this promise is realized, however, depends on the outcomes of three unpredictable processes.

The first is how FDA responds to the requests for additional studies, because they could further delay approval of this innovative product and discourage investment in related products. At the public meeting, critics expressed a variety of concerns about the agency’s proposed finding that the GE salmon was as safe as ordinary farm-raised Atlantic salmon. They claimed that the sample size for the safety analyses was inadequate, that levels of a growth hormone (IGF1) could be too high, and that the safety assessment was not conducted on the actual fish that will be marketed. Other speakers argued that concerns over sample size were overblown, because the various chemical analyses of the GE salmon found levels generally close to conventional salmon, and that the variance in levels among conventional salmon, indeed among commonly consumed fish of different species, was not well characterized in the scientific literature and not generally a safety concern.

FDA staff, displaying an unbecoming level of humility, did little to respond to these concerns. None of the criticisms made at the meeting however, constituted a cogent argument that approval would actually result in food that failed to meet the relevant standard of providing reasonable certainty of no harm when compared with farmed Atlantic salmon. Unless there is new work submitted to FDA beyond what was presented at the meetings, FDA should approve the product without requiring more studies.

The second issue is how to label food derived from AquAdvantage salmon. Given multiple surveys showing that consumers desire such labeling, should the government require the fish to be labeled as GE? The law governing food labeling does not give FDA authority to require companies to list the development method or production method on a food label simply based on consumer interest in having such information. Indeed, there is no mandatory labeling of food from GE crops for this reason. The agency  can, however, require labeling to reveal a “material difference” in a food from what would be expected for that food, such as the smell or taste of the fish;  the likelihood it causes an allergic reaction; or its functionality, for example, with respect to cooking methods.

But such differences do not exist in the case of AquAdvantage salmon so there will be no FDA-mandated labeling of food derived from it. The better public policy option is to enable nonmaterial information about a food, such as whether it meets kosher or halal restrictions, is raised using organic agricultural practices, or is developed using a specific technology, to be provided to consumers through voluntary labeling.

One approach to voluntary labeling that AquaBounty is probably considering is whether to sell AquAdvantage salmon under a brand name that permits its identification, for example, “fast-growing.” Although AquaBounty may consider such a voluntary agreement to label food products from its AquAdvantage salmon, it is unlikely to make a formal commitment to do so for all of such products indefinitely, given the likely FDA finding of no “material difference.”

Companies marketing food from conventional salmon may choose to add to their labels “absence” claims indicating that these products are not derived from GE animals, but such claims must be truthful and not misleading under current law. This means that they may not suggest that non-GE products are in any way superior to GE products and they must be supported by a factual basis, that is, some tracing or testing system indicating that the food is in fact not from a GE animal.  It is uncertain how food producers and distributors will choose to respond to the apparent demand for such information about whether salmon fillets are GE. And so there is substantial uncertainty about how fast the market for such products will grow.

The third and final question is how other governments will respond. Some countries, such as Chile, may see GE AquaBounty salmon as a threat—it will compete with traditional fish farms and decrease their revenue.  Others, perhaps in Europe, may contemplate bans on food from GE animals or, if these are not labeled, bans on all U.S. exports from a class of foods. Some, such as China, will continue investing in the development of new products using this technology, even as they regulate food safety in a manner less transparent and participatory than the process adopted by FDA. 

Aquabounty’s GE salmon is a small fish to carry the burden of demonstrating the promise of such technology.  FDA should expeditiously approve the product. Such a decision would set stakeholders in the food distribution and retailing industry and consumers in motion to adapt to a new world with superior salmon and other innovative GE products.

Randall Lutter played a key role developing FDA’s policy regarding GE animals when he served as FDA’s Deputy Commissioner for Policy from 2007 to 2009. ​​