The Food and Drug Administration’s public meetings last week on what may be the first genetically-engineered (GE) animal marketed for human consumption have stimulated a lot of discussion. People differ on whether the sale of AquAdvantage Salmon — which, due to the addition of a gene from the Chinook salmon, grows to full-size in less than half the time of its non-engineered Atlantic salmon cousins — will be a boon or a bane.
Would you eat this GE salmon? I would. There need not be any conflict here, though, as grocery stores are filled with foods that some people like and others shun. Why not allow people the freedom to make the decision for themselves? To answer this question, it’s important to understand how the FDA regulates products of biotechnology.
The majority of our food supply (the exception being food harvested in the wild) has been genetically modified over the years through selective breeding and other traditional methods. Recent concerns over “genetic engineering” relate to the newer techniques using recombinant DNA (rDNA). Many scientists find these concerns unfounded, because the new biotechnology techniques are more precise and reliable than traditional, trial-and-error techniques. For years the stated US policy has been to focus regulation on the characteristics of the product of genetic engineering, rather than the particular method used to achieve it. Following this philosophy, rDNA techniques have been applied successfully to key food crops, so that today over 70 percent of the corn and 90 percent of the soybeans grown in America are genetically engineered. To date, however, the U.S. government has never approved a GE animal for human consumption.
In January 2009, in a move that many saw as contrary to the long-standing policy of regulating based on the risks of the product, rather than the method, the FDA announced that it would regulate GE animals as if they were new animal drugs, requiring pre-marketing approval from FDA’s Center for Veterinary Medicine (CVM). The new animal drug approval process requires the manufacturer to provide the FDA with extensive data demonstrating both the “safety” and the “efficacy” of the product. In the case of the salmon, which AquaBounty has been developing for about 15 years, CVM’s preliminary determination is that it is “safe” for human consumption (because the salmon is indistinguishable from other Atlantic salmon) and the environment (the AquAdvantage salmon do not risk interbreeding with wild salmon because they are all sterile females and will be raised in land-based tanks). Indeed, it appears that the GE salmon will be better for the environment than either wild-caught salmon (which suffer from overfishing) or traditionally-farmed salmon (which may pose environmental risks). CVM has also determined the salmon is “effective” in that they really do grow faster.
If, as expected, FDA’s CVM does approve the GE salmon, a separate arm of FDA — its Center for Food Safety and Applied Nutrition (CFSAN) — will make a decision as to whether and how it should be labeled. Under FDA’s food labeling rules, the agency cannot require the salmon to carry a separate label unless it is materially different from non-engineered salmon (which it appears not to be). That makes sense to me, but here’s the rub: FDA rules may also not allow producers to inform consumers about whether the Atlantic salmon they’re purchasing involved the spliced Chinook gene. This is because FDA is not only concerned that labels be truthful, but that they not mislead consumers. In deciding whether firms selling conventional Atlantic salmon can label their products as not genetically engineered, the FDA will not only consider whether such claims can be supported (e.g., through tracing or testing) but also whether such a claim might falsely imply that the non-GE salmon is safer.
Prohibiting truthful voluntary labeling would be an unfortunate abridgement of commercial speech. While I will enthusiastically purchase and eat the new GE salmon (once FDA approves it), I respect the preferences of those who would rather not and of producers of non-GE salmon who would like to advertise the fact that their salmon is not genetically engineered. To meet the diversity in preferences, FDA should avoid a one-size-fits-all labeling prescription. At a minimum, it should do what it did in an analogous situation in 1994, when it allowed cows to be treated with recombinant bovine growth hormone (rBGH) to increase milk production. Then, it issued guidance detailing when and how labels could indicate that milk was produced from cows not treated with rBGH. (My all-natural Greek yogurt carries a label that says: “We oppose the use of rBGH. The farmers who supply our milk and cream pledge not to treat their cows with rBGH. The FDA has said no significant difference has been shown, and no test can now distinguish, between milk derived from rBGH-treated and untreated cows.”)
I hope FDA approves the GE salmon soon and then gets to work approving the enviro-pig (engineered to produce less pollution), the tuna-pig (high in omega-3 fatty acids) and cows resistant to mad cow disease.
Susan E. Dudley is Director of the GW Regulatory Studies Center and Research Professor in the Trachtenberg School of Public Policy and Public Administration of the George Washington University.