Opposing AquaBounty’s GE farmed salmon

RE: Docket No. FDA–2011-N-0899 Genetically Engineered Salmon
Submitted electronically

April 26, 2013

To whom it may concern:

Thank you for the opportunity to comment on Docket No. FDA-2011-N-0899 regarding AquaBounty’s AquAdvantage genetically engineered (GE) salmon. PCC Natural Markets is a natural foods cooperative with nine retail stores in the Seattle area and nearly $180 million in annual sales. On behalf of our 50,000 members, I strongly urge FDA to reject approval of this fish.

The process leading to FDA’s recommendation for approval of AquaBounty’s transgenic salmon should be rigorous and comprehensive. It has not been either. Much more data is needed to demonstrate safety.

Our member-owners have three major areas of concern: human health, the “New Animal Drug” designation and its implications for animal welfare, and environmental impacts.

If FDA does approve the salmon — despite widespread opposition by consumers and other stakeholders across the country — we request the fish at least be subject to mandatory labeling.

Human health

We primarily are concerned that the Environmental Assessment (EA) released by the FDA in 2010 found the salmon is safe to eat without adequately studying potential allergenicity and hormonal problems. Multiple testimonies by health, environment and animal welfare experts at the VMAC meeting on September 19-20, 2010 raised the following concerns:

I. Allergenicity

The potential of the AquAdvantage GE salmon to cause allergic reaction has not been studied adequately. FDA’s analysis of the GE salmon’s allergenic potency focused on a 2006 study furnished by AquaBounty itself. FDA has moved forward with approval based on tests of allergenicity of only six engineered fish.

According to Dr. Michael Hansen, senior scientist at Consumers Union, even this extremely limited sample showed an increase in allergy-causing potential.


Research shows GE salmon has raised levels of IGF-1, a hormone linked to prostate, breast, and colon cancers in humans. FDA’s analysis of IGF-1 levels focused on only two studies, a peer-reviewed publication from 1992 and an AquaBounty study from 2004. Scientists have found that foreign growth hormones in transgenic fish may increase production of other compounds, such as insulin, in the fish.

III. Other nutrients

GE salmon is less nutritious than normal (wild or farmed) salmon. The FDA’s review found significant differences between GE and non-GE salmon in vitamin B6 and other nutrients. FDA also found GE salmon has the lowest omega-3 to omega-6 ratio of all the salmon in the studies it reviewed. In its analysis of GE salmon’s nutritional content, the FDA depended on one dataset supplied by AquaBounty from 2003. This is insufficient.

Almost the entire basis of FDA’s analysis of food safety issues was based on data provided by AquaBounty, not independent, peer-reviewed research. Many experts agree the studies were poorly designed. Many critical data sets looked at only a handful of fish.

Further, there has been no safety testing of fish grown in Panama, where AquaBounty intends to raise the GE salmon. The health and safety of fish can be affected by growing conditions.

There should be more rigorous studies using better experimental design, with more sophisticated or sensitive methodology and a large enough sample size to perform an analysis to make reasonable conclusions.

“New Animal Drugs” and animal welfare

We do not believe GE salmon should be considered for approval as a “New Animal Drug.” This is an inappropriate designation.

FDA needs to design a new framework for evaluating the safety of the genetically engineered animals. This is a novel technology that should not be lumped into existing categories of FDA oversight.

FDA’s “New Animal Drug Approval” (NADA) process is designed for evaluation of new animal drugs, not genetically engineered animals. NADA was not designed to analyze the consequences of new animal drugs to human or environmental health.

Regulating GE animals as new animal drugs raises several concerns about animal welfare. First, the drug model is ill-suited for handling impacts to animal health and welfare. For example, using the drug model, lots that are found to be out of specification are destroyed. That makes sense when the lot destroyed is a batch of pills; it’s another when the “lot” is living animals.

Also, a drug typically is designed to provide some benefit to animal health, which FDA would weigh against potential risks. Genetic engineering of salmon does not benefit the fish in any way. The FDA has not indicated how it can make approval decisions for a drug that has no benefit but does carry risk of harm.

The NADA is required to show that a proposed drug, in this case the genetic engineering, is safe for the animals involved. We understand criticism that the data in Aquabounty’s NADA falls far short of meeting the animal safety requirement.

The animal health studies reviewed by FDA underreport, or fail to detect, problems in AquAdvantage salmon:

  1. Only healthy fish were studied. Fish with moderate to severe malformations and fish that were diseased, dying, or did not grow well were culled before any of the studies began. There is no data on these fish, so FDA’s conclusions rely on observations of only the healthiest fish (VMAC Briefing Packet, p 26, 27, 33, 36-37).
  2. Limited data were studied. Samples of just 12 fish were examined for only two weeks. It is not known how the genetic modification affects the fish at different life stages. The fish were not grown in Panama, where they’re supposed to be raised, or under production or approval conditions (VMAC Briefing Packet, p 24).
  3. There was no data on fitness. Almost none of the fitness studies FDA relies on in the Environmental Assessment (EA) involve the GE salmon under consideration. So there is no data on AquAdvantage salmon reproduction, feeding behavior, etc. (EA section 5.2.2).
  4. FDA’s conclusions are not supported. FDA admitted “it would be essentially difficult to draw any strong conclusions” from the animal health study but does so anyway in the EA and fails to mention any limitations or uncertainty in its findings.

Preliminary findings indicate GE salmon experience “increased frequency of skeletal malformations, and increased prevalence of jaw erosions and multisystemic focal inflammation” (VMAC Briefing Packet, p 45).


AquaBounty’s EA fails to address the real threats posed by escaped transgenic salmon on wild salmon populations and local ecosystems. It also ignores the environmental harm caused by raising and feeding these fish.

AquaBounty fails to consider the serious environmental impact of feeding farmed salmon — exacerbated by the fact that AquAdvantage Salmon is engineered to be faster growing. Salmon farming already consumes a significant amount of wild fish caught in the ocean, and the AquAdvantage salmon consumes up to five times more than its non-GE counterpart. This threatens many wild fish populations and their ecosystems.

No studies have been conducted on the feed efficiencies from recent lines of GE salmon. The only relevant data presented in the EA comes from one 1999 study conducted by researchers at the University of Manitoba [i] that did not look at feed efficiencies but rather at increased appetites. Also cited was a set of studies conducted in 2000 by employees of AquaBounty and researchers at the University of Prince Edward Island.[ii]

Even these limited data sets raise major concerns about the impact that feeding GE salmon at a commercial scale will have on broader marine environments and fisheries. According to the 1999 study, “AquAdvantage relatives consumed approximately five times more food than same-age controls that were also size-matched by delaying hatch time of the genetically engineered salmon: this consumption differential appears to derive from the increased feeding motivation of the GE salmon, which were 60% more likely than controls to be observed at both safe and risky foraging sites, and the increased willingness of the transgenic salmon to feed in the presence of a predator” [iii] (emphasis added).

The 2000 AquaBounty study showed better results but still found that GE salmon consumed 2.1 to 2.6 times more food than their non-GE counterparts [iv] and that “during pre-smolt growth these GE salmon consume much larger amounts of food than size-matched controls on a daily basis when fed to satiation three times per day under hatchery conditions.” [v] AquaBounty claims its fish have higher feed efficiencies than other farmed salmon as part of its claim that GE salmon are good for the environment, but this claim does not match the data provided.

Another concern is the likely escape of GE salmon from pens where they are raised. AquaBounty claims its GE fish areis sterile but admits the sterilization process does not work on as much as up to 5 percent of all eggs.

Hundreds of thousands of farmed salmon escape from contained systems every year due to damage by storms or wear and tear. AquaBounty’s assessment admits the facility where the eggs would be fertilized on Prince Edward Island in Canada is surrounded by an abundance of favorable habitats for fish species, including Atlantic salmon.

Research in the Proceedings of the National Academy of Sciences notes the escape of just 60 GE salmon into a wild population of 60,000 would lead to the extinction of the wild population in less than 40 fish generations [vi]. Faster growing GE fish that reach physical maturity faster will attract mates away from their natural counterparts, which can affect the reproductive success of wild species.

Under the conditions set forth in the EA, at most about 14,000 fish per year would be grown out in four tanks in Panama — hardly enough to provide seafood to a growing world population, service the aquaculture industry, or repay the millions of dollars invested by AquaBounty.

There is no way AquaBounty or the FDA can guarantee the eggs will be raised only in specific contained facilities on Prince Edward Island and Panama, as proposed in AquaBounty’s application. AquaBounty, in fact, admitted to VMAC at the meeting in September 2010 that it does plan on raising GE fish in America and other places.
A recent FOIA indicates the U.S. Department of Fish and Wildlife already has received applications to grow the fish in the U.S.

Given that the production scenario outlined in AquaBounty’s application does not represent the expected scenario for these fish, the EA cannot adequately address cumulative impacts to the environment, especially the global commons .

A more comprehensive EA must consider realistic production scenarios, including cases where containment measures are less than rigorous and escape is probable. The current EA fails to do so.

According to environmental analysis expert, Ann Kapuscinski of Dartmouth College, ensuring the fish are confined without the possibility of escape is crucial to prevent environmental harm. She says if physical confinement fails and sterile transgenic fish regularly escape into environments where they can thrive, it is important to realize they still could alter the environment.

In her testimony to VMAC at the September 2010 meeting, Kapuscinski raised two major concerns:

  1. How will the FDA assure and verify GE fish are confined at the two facilities (in Prince Edward Island and Panama) and in future facilities, as farming of these fish proliferates?

    “Confinement measures can appear rigorous but such complex safety systems are prone to human error and equipment failures,” she says. “The FDA should require a quantitative failure mode analysis for all the confinement methods. Failure mode analysis is standard practice for technology assessment. As commercial production of these fish proliferates, an even greater challenge is how to assure multiple confinement at many larger facilities in different environments and nations. Does the FDA have the resources and sufficient overseas jurisdiction for adequate surveillance?”
  2. The EA does not give the full information needed to predict environmental effects of AquAdvantage salmon. It stops at estimating that the likelihood of escape is “extremely small” due to multiple confinements at the two facilities. But this assumes 100 percent of the fish remain contained and none escape. Even if this happens, she says, it’s necessary to assess ecological consequences and then estimate the overall risk.

    “The assessment does not adequately address the major questions that should be asked about genetic and ecological risks,” she says. “Empirical studies have shown there is high scientific uncertainty in predicting overall fitness and ecological effects of transgenic fish because it is extremely challenging to extrapolate to nature from experiments using simulated natural conditions in the laboratory.”

In conclusion, approving GE salmon as the first genetically engineered animal for human consumption sets a precedent. The research to support that decision should be rigorous and transparent. So far, it has not been.

Independent, comprehensive research is needed to demonstrate safety of AquaBounty’s AquAdvantage transgenic salmon before it’s approved. FDA should consider how approving this salmon violates its mandate to protect public health.

FDA also should consider the utility of approving a fish that the vast majority of consumers reject. A new poll from Zogby shows 77 percent of Americans would not eat GE salmon if given a choice.

More than 2,500 grocers, including PCC Natural Markets, have pledged never to sell GE salmon. More than 40 members of Congress asked FDA to stop the approval process until concerns over risks, transparency and oversight are satisfied.

If you do approve the fish despite this widespread opposition, you must at least require it to be labeled to preserve freedom of choice by ensuring consumers an informed choice. .


Eli Penberthy
Public Affairs, PCC Natural Markets


[i] Abrahams, M.V. and A. Sutterlin (1999). “The foraging and antipredator behaviour of growth-enhanced transgenic Atlantic salmon.” Anim. Behav. 58: 933-942.

[ii] Cook, J.T., M.A. McNiven, and A.M. Sutterlin (2000a). “Metabolic rate of pre-smolt growth-enhanced transgenic Atlantic salmon (Salmo salar).” Aquaculture 188: 33-45; Cook, J.T., A.M. Sutterlin, and M.A. McNiven (2000b). “Effect of food deprivation on oxygen consumption and body composition of growth-enhanced transgenic Atlantic salmon (Salmo salar).” Aquaculture 188: 47-63.

[iii] Environmental Assessment for AquAdvantage® Salmon. Page 33.

[iv] Ibid, page 30.

[v] Ibid, page 92.

[vi] Robert H. Devlin et al., “Population Effects of Growth Hormone Transgenic Coho Salmon Depend on Food Availability and Genotype by Environment Interactions,” Proceedings of the National Academy of Sciences, vol. 101, no. 25, pp. 9303-9308, (June 22,2004), http://www.pnas.org/cgi/content/abstract/101/25/9303.

Related reading

Support Codex Alimentarius guidance on GE labeling

Sign-on letter to the USDA and FDA, urging support for Codex policy allowing countries to have different GE labeling policies.

Rulemaking for national GE labeling

We respond to USDA questions for stakeholders on what we want to see in GE labeling.

FDA proposes guidelines to regulate GE animals

Comments submitted to the U.S. Food and Drug Administration during public comment period on regulation of genetically modified animals.