Animal & Veterinary

AquAdvantage Salmon - Response to Public Comments on the Environmental Assessment

FDA’s Response to Public Comments

I. General Overview of Status

On December 26, 2012, the Food and Drug Administration (“FDA,” “the agency,” or “we”) released a Draft Environmental Assessment (“EA”) and preliminary Finding of No Significant Impact (“FONSI”) for public comment. The draft EA and preliminary FONSI were the agency’s preliminary analysis of likely significant impacts on the human environment of the United States of a potential FDA approval of a New Animal Drug Application (NADA) related to AquAdvantage Salmon, a genetically engineered Atlantic salmon, as required by the National Environmental Policy Act (NEPA) [42 USC § 4321 et seq] and FDA’s environmental impact considerations regulations [21 CFR Part 25.] These documents were open for public comment for 120 days; the comment period closed on April 26, 2013.

FDA’s Dockets Branch received and logged in comments according to the date of receipt. Once entered into the Docket’s system, each comment was forwarded to the staff of FDA’s Center for Veterinary Medicine (CVM or the Center) for review. In this document, we provide a general summary of public comments we received that are relevant to the draft EA and preliminary FONSI, and our response to those comments.

Comments are available for the public to review at regulations.gov docket number [FDA-N-2011-0899]. Alternatively, comments may be viewed at the Division of Dockets Management, 5630 Fishers Lane, Room 1061, HFA-305, Rockville, MD 20852, or by calling 301 827-6860 for assistance in finding a particular comment.

The agency thanks the public for the time and effort they put into providing these comments.

On November 19, 2015, FDA issued a final EA and final FONSI. Unless specified otherwise, references to the EA or FONSI refer to the final documents.

II. Overview of Comments

According to the Dockets Branch, the FDA Docket [FDA-N-2011-0899] logged a total of approximately 38,000 comments as of the close of the comment period. Most of these comments were form letters (what Dockets logged as a single comment under one submission may have contained multiple copies of the same form letter), or made general comments about how individuals felt about a potential approval of AquAdvantage Salmon1, or approvals of genetically engineered (GE) animals in general.

Most of the form letters and single-issue comments FDA received expressed general concern or opposition to the approval of AquAdvantage Salmon. Many of these comments indicated that the commentators had not carefully read the draft EA or preliminary FONSI, as their comments were based on the assumption that the AquAdvantage Salmon would be raised in net pens, and were concerned about escapes from those net pens. Some comments advised the agency to focus on commercial production of fast growing non-GE salmon or questioned the economic feasibility of commercial production of the AquAdvantage Salmon. Other comments expressed a favorable opinion about the AquAdvantage Salmon application noting the thoroughness of the scientific, risk-based review and the benefits of AquAdvantage Salmon for sustainable aquaculture.

Many comments also provided opinions on whether food from AquAdvantage Salmon or GE animals in general should be labeled to indicate whether it was derived from a GE animal. The FDA Center for Food Safety and Applied Nutrition has addressed comments concerning food labeling at http://www.fda.gov/downloads/Food/GuidanceRegulation/GuidanceDocumentsRegulatoryInformation/ LabelingNutrition/UCM469766.pdf.

Of the approximately 38,000 comments submitted to the Docket, the agency determined that approximately 90 specific comments from nearly 70 different organizations and individuals were responsive to the Federal Register notice requesting comments on the draft EA or preliminary FONSI and were substantive in nature. We considered the comments responsive because they specifically addressed components of the draft EA or preliminary FONSI, and substantive because they contained specific suggestions, criticisms, or positive substantiations of the agency’s analysis or conclusions. FDA has summarized and responded to these comments by topic in the body of this document.

Several of the substantive comments also included suggestions or criticisms not specifically relevant to the draft EA or preliminary FONSI, but addressed FDA’s preliminary review of data submitted by the sponsor in support of the NADA, as summarized in the Briefing Packet made available to CVM’s Veterinary Medicine Advisory Committee (VMAC) and that had been posted on our website at http://www.fda.gov/downloads/AdvisoryCommittees/CommitteesMeetingMaterials/ VeterinaryMedicineAdvisoryCommittee/UCM224762.pdf. Although some of these comments were substantive, they were not relevant to the request for public comments on the draft EA and preliminary FONSI and, therefore, are not addressed directly in our responses in this document. Because the public may have interest in some of the issues raised, however, we have addressed these comments in the accompanying Questions and Answers on the AquAdvantage Salmon Approval found on our website at http://www.fda.gov/AnimalVeterinary/DevelopmentApprovalProcess/GeneticEngineering/ GeneticallyEngineeredAnimals/ucm466221.htm, and in a separate document that addresses issues and comments raised in conjunction with the VMAC meeting on AquAdvantage Salmon.

Comments were submitted by individual scientists, individual technology providers, trade and professional associations, individual consumers, consumer and environmental advocacy groups, foreign governments or regulatory agencies, and members of U.S. Congress and state legislatures.

III. FDA’s Comment Summaries and Responses

The following is a compilation of the substantive comments on the draft EA and preliminary FONSI that FDA received during the public comment period that ended April 26, 2013. We have grouped these comments into two overarching groups: (1) those addressing legal, regulatory, or administrative issues, and (2) those addressing scientific or technical topics. In cases where there is some overlap between the two overarching categories, the comments are placed in the section that is most appropriate to the response.

A. Legal, Regulatory, and Administrative Issues

This broad category of comments includes those that specifically addressed the agency’s administrative handling of the draft EA and preliminary FONSI. Comments addressing legal, regulatory, or administrative issues not relevant to the EA/FONSI are addressed in the Questions and Answers or the Comments and Responses to the Veterinary Medicine Advisory Committee Meeting.

(1) Transparency

Comment Summary: Some comments indicated that the agency was insufficiently transparent in its dealings with the public by not releasing all of the data and information submitted for evaluation.

Response: The agency generally is not permitted to release information contained within an investigational new animal drug file, or even to confirm or deny its existence. 21 CFR 514.11. This regulation is based on the fact that such information is generally confidential commercial information. 21 CFR 25.50. However, with the sponsor’s permission, the agency made publicly available a significant amount of material during the course of its deliberations on the application. In addition, the agency held a VMAC meeting for which it provided the Committee and the public with highly detailed summaries of the data and information that it had been reviewing in order to obtain input from the Committee and the public. Further, the agency voluntarily made its draft environmental assessment and preliminary findings available for public comment, and is taking this opportunity to present the public with a summary of those comments and our responses. The Freedom of Information Summary, found at http://www.fda.gov/downloads/AnimalVeterinary/DevelopmentApprovalProcess/GeneticEngineering/ GeneticallyEngineeredAnimals/UCM466215.pdf provides a detailed analysis of the data and information used by the agency in making its decision to approve the NADA for AquAdvantage Salmon.

(2) Similarity of FDA’s draft EA and Sponsor’s EA

Comment Summary: Comments were submitting opining that the draft EA released for public comment in December, 2012 had almost no changes from an earlier version FDA released in conjunction with the VMAC meeting in 2010, and failed to address comments submitted to the VMAC.

Response: It is important to note that the 2010 draft EA released in conjunction with the VMAC meeting was prepared primarily by the sponsor’s consultant in consultation with scientists from the sponsor, AquaBounty Technologies (ABT) and, although the agency reviewed all drafts of the document, the 2010 draft EA contained the sponsor’s analysis and conclusions. That document contained information specific to the conditions present at the time that it was completed (August 25, 2010). Subsequent to that, the agency conducted its own analysis and, starting with the sponsor’s draft, prepared its own draft EA, which is considerably more complete. It included updated information on the practices at both ABT’s egg production and grow-out facilities, as well as the results of agency inspections and site visits, and interactions with other federal agencies. Although the fundamental containment approach has not changed (primary reliance on physical containment at the PEI breeding facility and multiple, redundant forms of containment, including physical, biological, geophysical, and geographic, at the grow-out facility in Panama), the scope and depth of FDA’s analysis in its draft EA are far more extensive than the sponsor’s draft EA, and include a more formalized risk analysis than that provided by the sponsor.

(3) Scope of EA

Comment Summary: Several comments stated that the scope of the EA was too narrow. These comments contended that the EA should include an analysis of the environmental impacts of current AquAdvantage Salmon facilities on Canada and Panama, of potential ABT facilities on the United States. and other countries, and of AquAdvantage Salmon grown in cages and net pens in open water. Some of these comments raised the same points in arguing that FDA must consider cumulative impacts of the present action in the EA. One comment stated that FDA should consider effects abroad under Executive Order 12114, Environmental Effects Abroad of Major Federal Actions.

Response: NEPA requires that FDA consider the environmental impacts of any “major federal action” that it takes. 42 U.S.C. § 4332(c). Approval of a new animal drug application is a “major federal action.” In this case, the major federal action is approval of a particular NADA for AquAdvantage Salmon. The NADA is for the specific conditions of use as described in the application and in the approval letter. These particular conditions of use allow production and grow-out of AquAdvantage Salmon at two facilities: one on Prince Edward Island in Canada, and one in Panama. As we state in the EA, because these facilities are outside the United States, and because NEPA does not require analysis of impacts in foreign sovereign countries, the EA considered environmental impacts in Canada and Panama only to the extent necessary to determine whether there would be significant effects on the environment in the United States due to exposure pathways originating from the facilities in Canada and Panama2. See Section 2.5 of the EA. Consistent with Executive Order 12114, FDA prepared an analysis of potential effects of this action on the environments of the global commons and on foreign nations not participating or otherwise involved in the action and determined that there would be no significant impacts. See http://www.fda.gov/AnimalVeterinary/DevelopmentApprovalProcess/GeneticEngineering/ GeneticallyEngineeredAnimals/ucm466350.htm.

FDA does not have jurisdiction to regulate potential facilities that would be located outside the United States that would produce fish for export to countries outside the United States. FDA also does not regulate products that are produced outside the United States and will never enter U.S. commerce. Although FDA does have jurisdiction to regulate facilities in the United States and products imported into the United States, FDA’s approval of the ABT NADA does not allow production and grow out of AquAdvantage Salmon in any facilities other than those in Canada and Panama. If ABT proposes to begin producing fish in the United States (or at additional locations outside of the United States with the intent to import food from them into the United States), it first must submit a supplemental NADA for the new production facilities, and this supplemental application will require its own NEPA analysis of potential environmental impacts of those facilities. See other discussions of supplemental NADA requirements in Section 2.1 and Section 4.1 of the EA. FDA approval of any such supplemental NADA must comply with NEPA requirements.3

At present, discussions of possible facilities in the United States are purely speculative. Although some commenters cite past statements by ABT officials about ABT’s intent to raise fish in U.S. facilities and ABT inquiries to the U.S. Fish and Wildlife Service (FWS) about importing live fish or eggs to the United States, at this time FDA has no other applications or proposals from ABT to develop and grow AquAdvantage Salmon anywhere but in the Canadian and Panamanian facilities covered by the current NADA. FDA could not analyze environmental impacts of such possible facilities without knowing specifics about them such as their location, the biological, hydrological, and geological features of the area, and any containment conditions that would be proposed. Specifics about a facility’s location, such as whether it is located near the ocean within the range of native Atlantic salmon, or inland and away from any water source that is a possible salmon habitat would have a significant influence on any analysis of potential environmental impacts.

The EA does not consider impacts of raising AquAdvantage Salmon in open water cages or net pens because the NADA did not propose raising fish in this manner and the approval does not permit raising them under such conditions. Moreover, FDA determined that net pens would not be an appropriate alternative due to potential escapes. The EA states, “[o]cean net pens or cages deployed in coastal marine locations have not proved to be consistently effective in preventing farmed salmon escapes to date, and likely would not insure sufficient primary physical/mechanical containment of AquAdvantage Salmon without further technological development, or significant decreases in the uncertainty associated with possible outcomes should AquAdvantage Salmon escape from ocean net pens in significant numbers.” Because use of net pens is not part of FDA’s action, the agency did not consider their use for AquAdvantage Salmon in the EA.

With respect to the issue of cumulative effects, as previously discussed in this Response, and stated in the EA, at this time, this is the only approval/action that the agency is considering. The Council for Environmental Quality’s NEPA regulations define cumulative impact as “the impact on the environment which results from the incremental impact of the present action when added to other past, present and reasonably foreseeable future actions . . . .” 40 CFR 1508.7. National Wildlife Federation v. FERC, 912 F.2d 1471, 1478 (D.C. Cir. 1990) (“[A]n EIS need not delve into the possible effects of a hypothetical project, but need only focus on the impact of the particular proposal at issue and other pending or recently approved proposals that might be connected to or act cumulatively with the proposal at issue.”); Society Hill Towers Owners' Ass'n v. Rendell, 210 F.3d 168, 182 (3d Cir. 2000) (future development identified in planning documents did not need to be considered in an EA for a hotel-parking garage project).

The analysis in FDA’s EA indicates that approval of the NADA for AquAdvantage Salmon would not have any significant environmental impacts on the United States. The absence of environmental impacts for the present action also means that there would be no “incremental impact” because this would be the first NADA approval for AquAdvantage Salmon and FDA is not aware of any specific reasonably foreseeable future actions on NADAs for GE fish at this time). As a result, there would be no cumulative impacts on the environment of the United States for the action to approve this NADA for AquAdvantage Salmon. Further, given that there are no other applications concerning AquAdvantage Salmon currently pending, and that any consideration of hypothetical locations and conditions is impracticable for the reasons explained in the previous portion of this response, there are no cumulative effects to consider. Should ABT submit additional applications to the agency, we would consider whether cumulative effects could occur and address them in accordance with applicable statutes and regulations.

(4) Potential for Raising AquAdvantage Salmon in Multiple Locations Requires an EIS

Comment Summary: Given that AquAdvantage Salmon could be raised in multiple locations that would not result in sales of edible products to the United States, the only way in which to determine the potential environmental risks would be to conduct a comprehensive EIS.

Response: As indicated in the previous response, the agency has no jurisdiction over facilities outside the United States that produce AquAdvantage Salmon solely for markets outside the United States. See, e.g., 21 U.S.C. 337a. There would, therefore, be no FDA action concerning these facilities. As such NEPA would not apply because there would be no major federal action with respect to any such facilities.

(5) Whether effects on the human environment are highly controversial

Comment Summary: A comment claimed that because FDA’s action in approving the NADA related to AquAdvantage Salmon is “controversial,” the agency is required to prepare an EIS. Citing case law, the comment stated that there is “a substantial dispute as to the size, nature, or effect” of FDA’s action.

Response: CEQ’s regulations list several factors to be utilized in determining the intensity of an action's impact on the human environment. Included in the list is the degree to which the effects on the quality of the human environment are likely to be highly controversial. 40 CFR 1508.27. The term “controversy” in the NEPA context refers to cases “where a substantial dispute exists as to the size, nature or effect of the major federal action rather than to the existence of opposition to a use.” Rucker v. Willis, 484 F.2d 158, 162 (4th Cir. 1973). See also Fund for Animals v. Frizzell, 530 F.2d 982, 988 (D.C. Cir. 1975) (“certainly something more is required besides the fact that some people may be highly agitated and be willing to go to court over the matter”). Moreover, “controversy is not decisive but is merely to be weighed in deciding what documents to prepare.” Hillsdale Envtl. Loss Prevention v. U.S. Army Corps of Eng’rs, 702 F.3d 1156, 1181 (10th Cir. 2012), quoting Town of Marshfield v. FAA, 552 F.3d 1, 5 (1st Cir. 2008). “When specialists express conflicting views, an agency must have discretion to rely on the reasonable opinions of its own qualified experts …” Marsh v. Oregon Natural Resources Council, 490 U.S. 360 (1989).

The comment cites “controversy” in two areas: the scope of FDA’s environmental review and the methodology of FDA’s review. First, the scope of FDA’s environmental analysis is appropriately limited to its action in approving the NADA related to AquAdvantage Salmon under the specific conditions of use described in the NADA and Letter of Approval. See discussion, section III A, comment 4 in this document.

Second, FDA is confident that the analysis it relied upon is appropriate. For any particular risk scenario, multiple risk assessment methodologies exist, each with their own merits, FDA has determined, based on its scientific judgment, the most appropriate method for its review. See discussion of data and analysis section III B, comment 1.

In contending that there is “controversy” regarding FDA’s methodology, a commenter focused particular criticism on FDA’s decision not to conduct a “quantitative failure mode analysis.” Failure mode analyses were developed by the U.S. Military in the 1940s and have subsequently been adopted by manufacturing sectors such as the automotive and aerospace industries.4 In effect, they are a step-by-step approach to estimating the amount of stress a particular component in a manufacturing system can take prior to “failing.” It becomes quantitative because it attempts to attach a failure rate to a particular component (e.g., one in 100,000 depressions on a brake pedal will result in leakage of a gasket or O-ring in a master or slave cylinder in the brake system.) This allows for the determination of the “weakest link” and for re-engineering to improve the performance of a particular component. In theory, failure mode analysis begins early in the conceptualizing of a manufacturing process so that faults can be identified and mitigated.

In the case of the containment systems for AquAdvantage Salmon, however, the sponsor did not choose to determine, for example, how many stresses a particular screen would have to endure prior to failing, but rather to account for the possibility that at least one of the components of the physical or biological systems would fail by the incorporation of multiple redundancies at each functional point (described in detail in the EA, Sections 2.6, 5.3.2, 5.4.3, 5.5.2, 7.2.1, 7.3.1, and 7.4.1) to ensure that the fish do not escape physical containment or establish in the environment.

Further, stressing the physical systems in Canada and Panama could result in the escape of AquAdvantage Salmon into the environment, which would be an undesirable outcome. Thus, in the case of the containment measures at PEI or Panama, the systems were not stressed to determine whether they would fail; rather the approach that was taken was to assume that any one component might fail, so therefore redundancy should be built into the system to compensate for any failures that might occur. Neither Canada nor the United States conducted a quantitative failure analysis (e.g., determining how often a screen might fail), but rather both Canada and the United States determined that there were sufficient redundancies in the containment that even if failures would occur, there would be sufficient compensating systems (redundancies) built into overall containment systems that, considered in toto, failure of any one system, even if likely, would not have a net effect on the containment of the system. Canada approached this from the perspective of a qualitative analysis that estimated the likelihood of failure and the degree of confidence that could be placed in that estimation. As in all assessments of risk, risks are multiplied so that if any one component becomes negligible, the entire risk scenario becomes negligible. See CSAS report summary, http://www.dfo-mpo.gc.ca/csas-sccs/Publications/ScR-RS/2013/2013_023-eng.pdf,disclaimer icon and EA, Section 2.5.1.

We note that, similar to FDA’s EA and FONSI, Canada’s independent environmental review concluded that the risk to the Canadian environment associated with manufacture and production of AquAdvantage Salmon is low with reasonable certainty. Additionally, FDA consulted with FWS and the National Marine Fisheries Service (NMFS) and shared its “no effect” determination with them. Although the commenter cites to questions that NMFS raised early in its review process, FDA subsequently met with NMFS, answered its questions, and, consequently, neither agency objected to FDA’s “no effect” determination. See EA, Appendix D.

(6) Supplemental NADA for Any New Facilities

Comment Summary: Some comments questioned whether ABT would have to notify FDA before opening new facilities and, if it did, whether FDA would review potential environmental impacts of such a facility.

Response: FDA regulations require that any “major” changes to an approved NADA require prior submission and approval of a supplemental NADA. 21 CR 514.8(b)(2). Such major changes include any change in the “controls, equipment, or facilities that has a substantial potential to have an adverse effect on the identity, strength, quality, purity, or potency of the drug as these factors may relate to the safety or effectiveness of the drug.” A new facility to produce AquAdvantage Salmon may affect its identity and quality. For example, a new egg production facility must have in place controls necessary to produce triploid, all female fish. Conditions at new grow-out facilities, including water quality and water temperature, may affect the growth and quality of fish. Additionally, we note that submission of a supplemental NADA for changes in manufacturing facilities is also a condition of the AquAdvantage Salmon NADA approval. Consequently, establishment of a new facility to produce AquAdvantage Salmon either in the United States, or in a foreign country to produce AquAdvantage Salmon for import into the United States, would require prior FDA approval of a supplemental NADA, which would trigger NEPA review requirements.

(7) Supplemental New Animal Drug Applications

Comment Summary: A concern was raised as to whether the supplemental NADA process is weaker than the original NADA process, and whether it would be subject to public comment in the event that the sponsor submits supplemental NADAs to cover changes made to any of the containment conditions that could affect the assessment of environmental risks.

Response: Supplemental NADAs are not “weaker” than NADAs. Guidance for Industry 82 (http://www.fda.gov/downloads/animalveterinary/guidancecomplianceenforcement/ guidanceforindustry/ucm052411.pdf) provides a general description of when a supplemental NADA would be required and, depending on the nature of the supplement, what kinds of data and information would be required to supplement an existing approval.

In Section 1 of the EA, we state that

“The sponsor must notify FDA about each proposed change in each condition established in an approved application, and obtain FDA approval of a supplemental application for the change where necessary. 21 CFR 514.8. Major and moderate changes to an approved NADA require the filing and review of a supplemental NADA. Approvals of such supplemental applications would constitute major agency actions and trigger additional environmental analyses under NEPA, unless otherwise excluded.”

As is the case with all new animal drug applications submitted to FDA, supplemental NADAs in and of themselves are not subject to public comment. Approval of a supplemental NADA, including such an application for changes to containment conditions would constitute an agency action that is subject to NEPA requirements.

(8) Consultation Under Section 7 of the Endangered Species Act

Comment Summary: Some comments stated that under Section 7(a) of the Endangered Species Act FDA must consult with FWS or NMFS concerning potential impacts on Gulf of Maine Atlantic salmon, an endangered species.

Response: Under section 7(a)(2) of the Endangered Species Act (ESA), 16 U.S.C. § 1531 et seq., federal agencies must ensure that actions they take are “not likely to jeopardize the continued existence of any endangered species or threatened species or result in the destruction or adverse modification of” a critical habitat of such species. Under the ESA and regulations implementing it, 50 CFR Part 402, agencies must determine whether their actions may affect an endangered or threatened species or critical habitat. If so, they must consult with FWS or NMFS. If the agency determines its actions will have no effect, no further consultation is required. With respect to AquAdvantage Salmon, FDA determined that an action approving the NADA would have no effect on endangered Gulf of Maine Atlantic salmon. FDA met with FWS and NMFS and shared its “no effect” determination with them. Neither agency objected to FDA’s determination. See EA, Appendix D. As a result, no further consultation with FWS or NMFS is required for this particular action.

(9) Scope of Alternatives

Comment Summary: Several comments asserted that the alternatives section of the EA was overly narrow and should have considered other alternatives such as review “and approval” by other agencies with expertise in fish biology such as NMFS, FWS, or EPA, non-GE methods of developing fast-growing salmon such as a method developed by SalmonBreed in Norway, and other conditions under an approval.

Response: Under the Council on Environmental Quality NEPA regulations, an EA must include “brief discussions of . . . the proposed action and alternatives.” 40 CFR 1508.9. Although the regulation does not describe the alternatives that the EA must discuss, the regulation defining an EIS states that an agency “must evaluate all reasonable alternatives.” 40 CFR 1502.14(a). Under the Food, Drug, and Cosmetic Act (FD&C Act), sponsors may submit NADAs to FDA. 21 U.S.C. 360b(b), and FDA must approve the NADA unless there are specific grounds under the FD&C Act to deny approval, in which case FDA must refuse to approve the NADA. 21 U.S.C. 360b(c), 360b(d).

FDA reviews the new animal drug that is the subject of an NADA under the conditions of use the sponsor describes in the NADA. In the AquAdvantage Salmon EA, the two alternatives that FDA considered were (1) the approval of the application under the proposed conditions of use, and (2) refusal to approve the application (the no action alternative). FDA also described other alternatives that involved different growing conditions and explained why these alternatives were rejected. Alternatives suggested by the commenters are not “reasonable” or feasible given that other agencies do not have the authority to review for approval an NADA under the FD&C Act and given that the application before FDA was for approval of a specific article in AquAdvantage Salmon and not for the development of fast-growing salmon by any possible method, many of which may not meet the definition of a new animal drug and may not be within FDA’s jurisdiction. We do note, however, that FDA met and discussed with NMFS and FWS potential environmental risks that could be associated with AquAdvantage Salmon.

Regarding the suggestion of implementing additional conditions under an approved application as an alternative action, some of the commenters’ suggested alternatives were covered in the EA. For example, a commenter suggested use of a closed containment system as an alternative. FDA considered and rejected this as a viable alternative because these systems “do not provide any significant advantage over flow-through systems.” See EA, Section 4.3. Other conditions suggested by a commenter regarding post-market recordkeeping, reporting, and monitoring are included in the AquAdvantage Salmon approved application (see Letter of Approval http://www.fda.gov/AnimalVeterinary/DevelopmentApprovalProcess/GeneticEngineering/ GeneticallyEngineeredAnimals/ucm466214.htm), either explicitly as individualized conditions in the approved application or as general requirements imposed by FDA regulations. Although we do not discuss these post-market recordkeeping, reporting, and monitoring requirements in the EA, we believe that they will reinforce the conditions under the approved application, which are evaluated in the EA. In its Letter of Approval FDA has clearly described the conditions under the approved application, including, among others, measures put into place to ensure that AquaBounty Technologies complies with conditions of use that are designed to prevent escape of eggs or fish.

(10) Economic and Social/Cultural Impacts

Comment Summary: Several comments stated that the EA should have considered economic, social, and cultural impacts of approval of the ABT NADA.

Response: As we note in the EA, FDA did not consider economic, social, and cultural impacts of its action because NEPA does not require consideration of such impacts if an EA concludes that the proposed agency action will not have an impact on the physical environment. The trigger that requires preparation of an EIS is a finding that an agency action will have a significant impact on the physical environment, not a finding of a social or economic impact. 40 CFR 1508.14; see Olmstead Citizens for a Better Community v. U.S., 793 F.2d 201 (8th Cir. 1986) (“an impact statement generally should be necessary only when the federal action poses a threat to the physical resources of the area….”). See also Metropolitan Edison Co. v. People Against Nuclear Energy, 460 U.S. 766, 773-76 (1983). Because a finding of a significant impact on the physical environment is a condition precedent to further analysis of impacts in an EIS and because FDA has determined that its action in approving the ABT NADA under the specified conditions of use will not affect the physical environment of the United States, further analysis of impacts, including socioeconomic impacts, in an EIS was not required.

(11) Executive Order on Environmental Justice

Comment Summary: A comment stated that FDA must comply with Executive Order 12898, Federal Actions to Address Environmental Justice in Minority Populations.

Response: Executive Order 12898 directs federal agencies to address disproportionately high adverse human health or environmental effects of their actions on minority and low-income populations. FDA’s action in approving the AquAdvantage Salmon NADA will not have disproportionately high health or environmental effects on minority or low-income populations. Some commenters raised the possibility of impacts on western and northwestern native-American tribes in whose culture wild salmon play an important role. FDA’s action will not have an impact on such tribes because the conditions under the approved application include that AquAdvantage Salmon will be developed and grown in facilities in Prince Edward Island, Canada and in Panama. The multiple, redundant physical containment measures as well as geographical containment, as described in the EA, ensure that there is virtually no possibility of AquAdvantage Salmon escaping these facilities and swimming to the regions of the western United States with Pacific wild salmon populations. Furthermore, as described in Section 7.4.1.6 of the EA, even if AquAdvantage Salmon could somehow migrate to the Pacific Ocean, there is no reasonable potential for hybridization between escaped AquAdvantage Salmon and native Pacific salmon, which are of a different genus, Oncorhynchus. Farm-raised Atlantic salmon on the west coast of the United States and Canada that have escaped in the Pacific Ocean have not interbred with wild Pacific salmon and, to date, there has been no compelling evidence of any colonization and establishment (i.e., self-sustaining populations) of Atlantic salmon in these areas.

(12) Executive Order Requiring Consultation and Coordination With Indian Tribal Governments

Comment Summary: Some commenters raised concerns about Indian Tribal rights with respect to wild salmon. In particular, a comment stated that under Executive Order 13175, FDA must consult with Indian tribal governments before approving the NADA related to AquAdvantage Salmon and another comment cited tribal rights with respect to wild Pacific salmon in the Klamath River Basin. One comment cited the Penobscot Indian Nation in Maine and its right to subsistence fish for Atlantic salmon in particular, and a comment by the Yurok Indian Tribe notes the Tribe’s rights in relation to wild, native Klamath River salmon in California.

Response: Among its provisions, Executive Order 13175 states that “each agency shall have an accountable process to ensure meaningful and timely input by tribal officials in the development of regulatory policies that have tribal implications.” The ExecutiveOrder defines “regulatory policies that have tribal implications as “regulations, legislative comments or proposed legislation, and other policy statements or actions that have substantial direct effects on one or more Indian tribes, on the relationship between the Federal Government and Indian tribes, or on the distribution of power and responsibilities between the Federal Government and Indian tribes.”

FDA’s action in approving the NADA related to AquAdvantage Salmon will not have a “substantial direct effect” on an Indian tribe. The EA notes that FDA’s approval of the NADA will have no effect on Gulf of Maine Atlantic salmon. As a result, this action will not impact Indian tribes that subsistence fish for Atlantic salmon in the Gulf of Maine. Additionally, as explained in response 11 above, FDA’s action will not have an effect on wild, Pacific salmon and, therefore, will not impact Klamath River salmon.

(13) Authority for Environmental Review

Comment Summary: Some comments reflected confusion about the legal authority underlying the EA.

Response: NEPA requires that FDA consider the environmental impacts of any major action it takes. In this case, the major action triggering NEPA review is FDA approval of the ABT NADA for a particular article in GE salmon, grown under the specific conditions described in the NADA and the approval letter. To meet its NEPA obligations, FDA prepared the EA to determine whether this action may result in a significant impact on the environment of the United States. Because FDA determined that its action would not have such an impact, the agency is issuing a Finding of No Significant Impact (FONSI). The EA and FONSI together fulfill the agency’s obligations under NEPA.

(14) Programmatic EIS for GE Animals Generally

Comment Summary: Some comments maintained that FDA must prepare a programmatic EIS for “the federal government’s broader program for the establishment and commercialization of GE animals” or for GE fish generally.

Response: Under CEQ’s NEPA regulations, a programmatic EIS may be prepared, and is sometimes required, “for broad Federal actions such as the adoption of new agency programs or regulations.” 40 CFR 1502.4. See also National Wildlife Federation v. Appalachian Regional Com., 677 F.2d 883, 889 (D.C. Cir. 1981) (“Two considerations are especially helpful in reviewing the responsible officials' decision not to prepare a programmatic EIS: (a) Could the programmatic EIS be sufficiently forward looking to contribute to the decisionmakers' basic planning of the overall program? and, (b) Does the decisionmaker purport to "segment" the overall program, thereby unreasonably constricting the scope of primordial environmental evaluation?”).

As described in CVM Guidance for Industry (GFI) 187, an inserted rDNA construct that is intended to affect the structure or function of an animal meets the definition of a drug under the FD&C Act. Products that meet this definition, and that are new animal drugs, are subject to applicable FD&C Act requirements concerning premarket approval. The approval of the NADA related to AquAdvantage Salmon is an individual action, just as approval of any other NADA is an individual action, with its own particular, potential environmental risks. FDA has no program “for the establishment and commercialization of GE animals” separate from the general NADA approval program established by statute. 21 U.S.C. 360b. Sponsors are developing many types of GE animals, some of which would be intended to be maintained under contained and highly controlled conditions whereas others would not. For example, the first NADA related to a GE animal that FDA approved was for a goat that produces a human biological drug. See 21 CFR 528.1070, Bc6 recombinant deoxyribonucleic acid construct. This species of animal and the conditions under which it is raised are significantly different in every respect from AquAdvantage Salmon. If FDA were to prepare a programmatic EIS prior to receiving NADAs for such potential products, the EIS would necessarily be under-inclusive because FDA cannot anticipate every GE animal for which it might receive a related NADA and the conditions and locations under which such animals would be developed and raised. This is true even with respect to fish generally. There are at least 100 species of finfish used in aquaculture worldwide, with growth conditions that vary with respect to water characteristics (e.g., salinity, temperature, clarity), diet (e.g., plant, fish meal, or some combination), and containment (e.g., ponds with little or no water exchange, constant flow-through, or recirculating tanks). Because FDA cannot anticipate every species of GE fish for which it might receive an NADA or every type of condition under which any GE fish could be grown, a programmatic EIS is neither required, nor, more importantly, would it be advisable or practicable given the multiple variables involved.

(15) Report in Consultation With NMFS

Comment Summary: Some comments noted that Section 1007 of the Food and Drug Administration Amendments Act of 1997, P.L. 110-85, requires FDA to consult with NMFS “to produce a report on any environmental risks associated with genetically engineered seafood products, including the impact on wild fish stocks.”

Response: Because AquAdvantage Salmon is the only GE seafood product that FDA has considered and for which sufficient data and information concerning environmental risks exist to enable meaningful analysis, the only report that FDA can issue on environmental risks that might be associated with “genetically engineered seafood products” (which FDA interprets to mean live fish) is one that addresses specific issues associated with AquAdvantage Salmon. FDA has no specific information about other such seafood products, including the conditions under which they might be raised. As a result, any report associated with all genetically engineered seafood products and their potential impact(s) on wild fish stocks would be so fraught with uncertainties regarding both the nature of the traits that had been genetically engineered into the “seafood products,” and their potential impacts, as to be impracticable. Therefore, the final EA and FONSI, which evaluate the environmental impacts of AquAdvantage Salmon under the conditions specified in the EA, constitute the report required under Section 1007 of the Food and Drug Administration Amendments Act (FDAAA).

(16) NASCO Williamsburg Declaration

Comment Summary: Some commenters pointed out that the United States is a signatory to the North Atlantic Salmon Conservation Organization (NASCO) and suggested that this precludes FDA’s approval of the NADA related to AquAdvantage Salmon.

Response: FDA’s participation in the North Atlantic Salmon Conservation Organization does not affect FDA’s approval of the AquAdvantage Salmon NADA. NASCO is an international organization whose members include six government parties, including Canada and the United States. NASCO’s stated mission is “to conserve, restore, enhance and rationally manage Atlantic salmon through international cooperation.” NASCO’s Williamsburg Declaration is a non-binding resolution adopted by its members to “Minimise [sic] Impacts from Aquaculture, Introductions and Transfers, and Transgenics on the Wild Salmon Stocks.” The Declaration states, “In view of the current lack of scientific knowledge on the impact of transgenic salmonids on wild salmon stocks, the use of transgenic salmonids should be considered a high-risk activity. There should be a strong presumption against any such use.” The Declaration also states that the parties “should apply the Guidelines for Action on Transgenic Salmon.”
These guidelines are not binding on NASCO members; however, they state that the parties should, among other actions, “take all possible actions to ensure that the use of transgenic salmon, in any part of the NASCO Convention Area, is confined to secure, self-contained, land -based facilities.” FDA believes that the two facilities that will be used for production and grow-out of AquAdvantage Salmon follow this recommendation in the NASCO Guidelines in that they are secure, self-contained, land-based facilities (see Section 5.4, 5.5 and 7.2 of the EA for additional information on these facilities and containment).

(17) Consideration of Certain State Laws

Comment Summary: A comment stated that the EA should consider state laws regulating GE fish, citing in particular California and Alaska statutes.

Response: Alaska state law requires that “genetically modified fish” be “conspicuously labeled to identify the fish or fish product as a genetically modified fish or fish product.” Alaska Stat. § 17.20(a)(14). Additionally, the Alaska legislature passed a resolution opposing AquAdvantage Salmon. In California, state law requires a permit for “importation, possession, transportation or rearing of, or research on transgenic aquatic animals.” Cal. Code Regs., tit. 14, § 671(c)(15).

Neither of these states’ statutes will affect the environmental impact of FDA’s action in approving the NADA related to AquAdvantage Salmon. The Alaska labeling requirement relates to labeling of food products, and the Alaska resolution does not impose any legal requirements. California’s regulations do not impact AquAdvantage Salmon because, under the conditions of the approval, AquAdvantage Salmon will not be developed or raised in California or any other locations except those specified in the application.

(18) Impact of EA on Canada’s and Panama’s Review Processes

Comment Summary: Some comments maintained that FDA’s EA had “prejudged” the outcome of Canadian and Panamanian environmental review processes.

Response: The EA assesses the potential impacts on the U.S. environment of approving the NADA related to AquAdvantage Salmon. FDA’s analysis is required by NEPA, a U.S. statute, and is not contingent on the reviews of any other countries.

We note, however, that Canada has completed an environmental analysis as part of its New Substances Notification Regulations [see Section 2.5.1 of the EA) for a change in production status of the PEI facility from research and development to commercial production. See CSAS report summary, http://www.dfo-mpo.gc.ca/csas-sccs/Publications/ScR-RS/2013/2013_023-eng.pdf.disclaimer icon The risk assessment associated with this analysis concluded that the risk to the Canadian environment associated with manufacture and production of AquAdvantage Salmon is low with reasonable certainty. We also note that Panama’s National Authority of the Environment has found that ABT’s facility “is complying with all the environmental commitments established.” See EA, Appendix G.

(19) Characterizing the Product for Commercial Sale and Distribution

Comment Summary: One comment pointed to language in the draft EA that suggested FDA was approving eyed eggs for sale (“AquAdvantage Salmon would be produced as triploid, all-female populations with eyed-eggs as the product for commercial sale and distribution.”).

Response: The language the commenter cites mischaracterized FDA’s action. FDA has clarified this language to reflect the correct product description and conditions under the approved application, which allow shipment of eyed eggs only from ABT’s facility in Canada to its facility in Panama to help avoid such confusion in the future (See Section 4.1.1 of the EA). FDA does not have jurisdiction over sales of eyed-eggs for production of AquaAdvantage Salmon outside of the United States that is not imported into the United States.

(20) World Trade Organization Rules

Comment Summary: A comment asserted that FDA’s authority to impose case-by-case requirements is questionable because they are likely to be vulnerable to challenge under WTO rules unless they are based on relevant international standards and are non-discriminatory.

Response: We do not believe that case-by-case conditions under approved applications are necessarily discriminatory and expect any such conditions to comply with U.S. trade obligations.

(21) FDA’s Ability to Enforce Conditions Under the Approved Application

Comment Summary: Several comments asked how FDA would enforce the conditions in the approval, raising concern that FDA may not be able to regularly inspect the facilities, in part because of the distance between FDA headquarters and the sponsor’s facilities (especially the Panama facility).

Response: Under the FD&C Act, a new animal drug that does not comply with its approved application is “unsafe” and an unsafe new animal drug is “adulterated.” 21 U.S.C. §§ 360b(a); 351(a).

The FD&C Act also deems adulterated any food that contains an unsafe new animal drug or a “conversion product thereof.” 21 U.S.C. 342(a)(2)(C)(ii). FDA may take enforcement action against adulterated drugs and foods, including refusing admission to imported foods and drugs that appear to be adulterated. If AquAdvantage Salmon were not in compliance with the conditions under the approved NADA it would be adulterated under these provisions and subject to FDA’s enforcement action. FDA will monitor for compliance with the conditions specified under the approved NADA. FDA’s ability to perform inspections of the Panama facility will be augmented by the location of FDA’s Latin American Office in neighboring Costa Rica. See also the Comment Summary and Response to Post-Approval Oversight in this document.

(22) Enforceability of Mitigations

Comment Summary: Some comments emphasized that mitigation measures must be enforceable.

Response: Although comments referred to “mitigations” that FDA is imposing, there are none. Mitigations are measures that are identified by using “the NEPA process to identify and assess the reasonable alternatives to proposed actions that will avoid or minimize adverse effects of these actions ....” 40 CFR 1500.2(e). In the case of the AquAdvantage Salmon NADA, the action subject to NEPA review is approval of the NADA, including the conditions as proposed in the NADA and as included as conditions of the FDA approval. In the EA, FDA determined that the action, under the conditions proposed, would not have a significant impact on the human environment of the United States. As a result, FDA did not require mitigations as “alternatives to [the] proposed action” in order to reach a FONSI. Nevertheless, as described in the Letter of Approval, http://www.fda.gov/AnimalVeterinary/DevelopmentApprovalProcess/GeneticEngineering/ GeneticallyEngineeredAnimals/ucm466214.htm, FDA will enforce the conditions under the approved application which are the focus of the comments, through ongoing reporting and recordkeeping requirements and inspections. Moreover, we note that FDA has inspected the PEI facility twice, and engaged in an extensive site visit to the Panama facilities. As a result, the agency is able to confirm that the conditions as described in the approved application are in place, the facilities are as described, and ABT has developed a set of operating procedures at each location to ensure the integrity of physical containment. Further, the Canadian government conducts periodic inspections of the PEI facility and has recently approved a change in the status of the PEI facility from a “research” to “commercial” facility based on a qualitative Failure Mode Analysis and extensive risk assessment.

B. Scientific and Technical Issues

This broad category of comment summaries and agency responses covers issues associated with the nature of data and information evaluated by the agency, FDA’s analyses of those data and that information, issues relevant to forms of containment, and other substantive issues associated with the EA.

(1) Uncertainty and Gathering Further Data

Comment Summary: Some comments said that FDA must gather further data due to uncertainty.

Response: Comments cited studies concerning transgenic fish in general in support of the assertion that uncertainty requires gathering additional data. As explained below and in the response to question (11) in this section, hypotheses regarding genetically engineered fish in general are non-specific and lack relevance to the particular conditions under which AquAdvantage Salmon will be developed and grown. For example, some comments raised the “Trojan gene hypothesis,” which explored possible extinction of the laboratory model fish medaka populations through the flow of a gene that confers a reproductive advantage while also rendering offspring of those medaka less able to survive in the natural environment. EA, Section 7.5.1.1.1. As cited by the originator of that hypothesis, however, when applied to AquAdvantage Salmon, this is not the case. See EA, Section 7.5.1.1.1 (“ …the data conclusively shows that there is no Trojan Gene effect as expected. The data, in fact, suggest that the transgene will be purged by natural selection.”; “the long-term risk of GE salmon is close to zero as no fitness advantages in any component were demonstrated, resulting in a purge scenario for the transgene.”). The commenters also inappropriately cite statements from the draft EA as evidence of uncertainty. For example, a comment cited the statement in the draft EA that “there are no specific data demonstrating that triploid AquAdvantage Salmon are indeed sterile….” This statement is followed by an extensive discussion of the scientific evidence that triploidy does indeed result in effective reproductive sterility in Atlantic salmon and other fish (Draft EA, Section 7.2.1.1.). In addition, the use of gynogenesis to produce an all female population provides an additional level of redundancy in biological containment.

It is important to note that there will always be some level of uncertainty surrounding any risk estimation, including that made specifically for AquAdvantage Salmon. One reason is that, as with all fish, the phenotype of AquAdvantage Salmon is influenced not only by their genotype, but by the particular environment in which they are found. For example, the phenotype of fish raised in tanks in a laboratory will differ from that of fish that live in the wild, even if they share identical genomes. This phenomenon is referred to as a genome by environment (G X E) interaction. Further, because to date, there have been no approvals for AquAdvantage Salmon, or free releases of AquAdvantage Salmon into the environment, risk assessors and regulators must extrapolate from studies conducted under laboratory or simulated natural conditions, including mesocosms or simulated streams. Generating additional data in laboratory studies or even mesocosms could somewhat reduce but not eliminate uncertainty due to all of the uncontrolled conditions in the environment. The analysis in the EA focuses first and foremost on the adequacy and redundancy of containment, as the conditions of the approval specifically require extensive and redundant physical and biological containment.

(2) Containment

Multiple comments expressed concern regarding particular forms of containment. Although the agency has attempted to address these individual concerns in the comments below, FDA emphasizes that the purpose of having a “containment system” consisting of multiple, redundant containment measures is to provide overlap of individual components to help ensure the integrity of the system as a whole. Containment therefore should be considered as the sum of the individual measures rather than any measure in isolation. Further, many of the comments regarding loss of physical or biological containment are based on examples of failures of net pen systems; the conditions under the approved application specifically preclude ocean net pens, and require that all fish, regardless of whether they are broodstock or intended for food production, are raised in contained, land-based facilities.

(a) General Containment Issues

Comment Summary: Several comments expressed the opinion that AquAdvantage Salmon would be allowed to be raised in net pens under this approval, or would eventually be raised in net pens. Another comment indicated that there are strong ecological reasons for abandoning open net-pen production and requiring movement to closed, land-based contained systems, while noting the resistance to doing so based on cost.

Response: The conditions under this approval for AquAdvantage Salmon are for rearing and grow-out only in physically-contained, land-based freshwater culture facilities, which are described in Sections 5.4 and 5.5 of the EA. Rearing in conventional cages or net pens is not permitted under this approval, and this is required to be indicated on the AquAdvantage Salmon shipping label. At this time, FDA agrees that there are good reasons to raise AquAdvantage Salmon in highly-contained, land-based systems, as described in the application and EA. Although almost all recirculating aquaculture systems have some water discharge during normal operations, and more extensive discharge during cleaning and maintenance, and thus are not totally closed, the redundant forms of containment required as conditions under the AquAdvantage Salmon approval provide adequate assurance that, even during discharge, any life stage of GE salmon would continue to be physicially contained.

(b) Physical Containment

Comment Summary: Several comments questioned the adequacy of the physical containment at the PEI and Panama facilities, stating that land-based facilities are not without risk and that multiple levels of containment could still be subject to human error and deterioration.

Response: ABT has provided FDA with detailed schematics and operating practices to indicate that the physical containment systems in the PEI breeding facility and the Panama grow-out facilities are more than adequate to provide physical containment of all of the life stages of AquAdvantage Salmon (see EA, Sections 5.4 (PEI) and 5.5 (Panama)). ABT has put into place security measures at both facilities to prevent unauthorized access, control movement of authorized personnel, and prevent access by predators. In addition, there are Standard Operating Procedures (SOPs) in place that describe the maintenance of all physical containment measures, as well as every other significant activity that occurs at each facility. FDA has verified and validated these SOPs during inspections and site visits.

Further, the containment programs reflect the principles outlined in the U.S. Department of Agriculture’s Agricultural Biotechnology Research Advisory Committee’s (ABRAC) Performance Standards for safely conducting research with genetically modified fish and shellfish (ABRAC Standards, 1995), as described in Section 3.2.1 of the EA. Because of the virtual impossibility of recapturing escaped fish, it is important that containment systems be redundant: ABRAC recommends that although the number of independent containment measures (i.e., physical, chemical, mechanical, and biological barriers) is site- and project-specific, they should generally range from three to five. ABRAC Standards at 39. The containment systems at both facilities meet or exceed this recommendation.

For example, in Panama, as described in the EA in Section 5.5, there are a minimum of 11 sequential physical barriers in place between the fry tanks and the local river; seven of these barriers are installed in the outflow from the grow-out tanks. With respect to netting, the grow-out tanks are sealed horizontally and vertically inside a cage comprised of netting supported by a rigid metal structure, which would make escape by jumping or by predator capture virtually impossible. In addition, the tanks are completely surrounded by chain link fencing topped with barbed wire, which would further preclude any entrance by predators. At the breeding facility in PEI, there are three major types of containment, which have multiple features that are intended to provide physical barriers for the larger fish, or to kill any escaped eggs or very small fish in the unlikely event that the physical barriers are overcome (see Table 2 and Figure 6 in Section 5.4.3 of the EA).

FDA has validated and verified the presence and operational integrity of these systems with two formal inspections of the PEI facility (one in October 2008, and the second in June 2012). We have also reviewed daily maintenance records made by the facility personnel. These records note, for example, that from time to time some of the larger fish at PEI do jump out of the tanks, but are found dead or moribund on the dry floor of the building that houses the tanks. The fish are many times larger than the very small (1.5 mm) openings in the drains in the floor of that portion of the facility, thus it is impossible for them to go anywhere. But even if they somehow did manage to get into the drainage system, this system has several additional screens further downstream with very small openings that would not let the fish pass through (see Figure 6 in EA).

With respect to the PEI nursery, where eggs are fertilized, triploided, hatched, and where very small fish are raised, we have carefully inspected (as has the Canadian government) the physical containment (described in detail in Section 5.3 of the EA) and found that several types of fine mesh screens made of plastic netting or stainless steel are present in the drain lines, fully functional, and sufficient to prevent any life stage of salmon from passing through and escaping the contained tanks. Chlorine pucks are also used in the drains during the times when fish are spawned so that, in the event that any fish embryos accidentally escape containment , they would be killed by the chlorine.

ABT has instituted a series of SOPs to perform frequent checks of the integrity of the mechanical systems that make up the physical containment of the facility, and has processes in place to ensure that the containment maintains integrity. Our inspection of these records indicates that the company has been diligent in adhering to those protocols. Nonetheless, these records, as well as the physical status of the containment systems, will be part of the focus of subsequent inspections by U.S. inspectors. In addition, Canadian authorities conduct inspections of the PEI facility. Since 2011, Environment Canada enforcement officers have conducted annual inspections of the PEI facility in order to verify compliance with the physical containment requirements. No violations of these requirements have been observed during these inspections.

Further, the Canadian government has issued a notification under its New Substances Notification Regulations (NSNR) (Organisms) to allow a change in production status from research and development to commercial manufacture. In order to make this determination, Canada’s Department of Fisheries and Oceans (DFO) conducted environmental and indirect human health risk assessments of AquAdvantage Salmon, which were then peer reviewed. CSAS report summary, http://www.dfo-mpo.gc.ca/csas-sccs/Publications/ScR-RS/2013/2013_023-eng.pdf.disclaimer icon

Because risk is a function of both hazard and exposure, the part of the Canadian risk assessment that addressed exposure (i.e., unintentional release of fish) included a qualitative Failure Mode Analysis to provide insight into the efficacy of all barriers and all operational procedures involving containment. This analysis considered (1) the potential for an acute failure of physical containment; (2) the likelihood of chronic release of fish at both the PEI and Panamanian facilities; and (3) the likelihood of release while in transit between the PEI and Panamanian facilities. DFO concluded that for the use scenario specified in the regulatory submission, which are identical to the conditions of use specified in the NADA submitted to FDA, exposure of AquAdvantage Salmon to the Canadian environment is expected to be negligible with a reasonable certainty. CSAS report summary at 12, http://www.dfo-mpo.gc.ca/csas-sccs/Publications/ScR-RS/2013/2013_023-eng.pdf.disclaimer icon

(c) Biological Containment

i. Adequacy of biological containment

Comment Summary: Concern was expressed in several comments that the PEI egg production facility would be maintaining fertile, diploid GE fish and that conditions of biological and geographical containment would not apply to these fish.

Response: It is correct that there is no biological containment for the AquAdvantage Salmon broodstock in the PEI egg production facility, as inducing sterility to broodstock is contrary to their intended use. As described in Section 5.3.1 of the EA, the broodstock are fertile and diploid and must remain so in order for the propagation of AquAdvantage Salmon. The EA recognizes that these diploid GE Atlantic salmon (referred to as ABT salmon) broodstock represent the greatest potential risk to the environment in the unlikely event of an escape. Potential risks posed by the ABT broodstock are discussed at length in Section 7.5.1.1.1 of the EA. FDA has determined that it is highly unlikely that these fish would be able to reproduce and establish in the local environment or farther afield in the event of an escape for a number of reasons. First, due to space limitations the number of adult broodstock in the PEI facility will be limited to several thousand at any one time; therefore, the potential for the mass release of many thousands to hundreds of thousands of post-smolt fish, as sometimes occurs during net-pen farming of Atlantic salmon, will not occur. Second, approximately half of the adult “males” in the facility are actually neomales (sex-reversed females) that generally cannot spawn on their own due to lack of functional sperm ducts. Third, non-triploid relatives of AquAdvantage Salmon are outperformed by their non-transgenic counterparts in most measures of reproductive performance and show reduced fitness compared to wild Atlantic salmon. See EA, Section 7.5.1.1.1. Fourth, Atlantic salmon are no longer found in any of the watersheds near the PEI facility despite repeated attempts to stock them there in the past. The lack of wild Atlantic salmon nearby not only lessens the possibility of any reproductive interactions with other Atlantic salmon (which do not exist in these waters), but perhaps more importantly, indicates that any very unlikely reproduction of AquAdvantage Salmon amongst themselves is unlikely to result in self-sustaining populations in this or surrounding areas.

ii. Effectiveness of the Induction of Triploidy

Comment Summary: Several comments noted that triploidy induction is not an efficient and reliable process. A number of comments emphasized that up to 5% of all eggs may be fertile and may significantly impact the environment in the event some of the fish in Panama escape.

Response: The efficacy of triploid induction is discussed in Sections 5.3.2.2 and 7.4.1.1 of the EA. It is highly efficient and reproducible from batch to batch, with an average effectiveness of 99.8% shown by two different method validation studies. Commercial release specifications, under the conditions of the approved application, require that, based on testing, at least 95% of the eggs be triploid (based on the statistical 95% lower confidence limit) or the entire batch of eggs must be destroyed. See EA, Section 7.4.1.2. For a batch of 10,000 eggs, this means that up to 500 eggs might not be triploid. FDA does not believe this number represents a significant risk for a number of reasons. First, the primary form of containment in Panama is physical containment, which is highly redundant and effective. Second, the most important form of biological containment is the fact that all AquAdvantage Salmon in Panama are females. Because all of these fish are female and there are no native or introduced Atlantic salmon in Panama (and no other related fish species which can reproduce with Atlantic salmon), no reproduction of AquAdvantage Salmon is possible even if these fish are not triploid. Triploidy is thus a secondary form of biological containment that does not need to be 100% effective in this case.

iii. Fertility Status of Triploid AquAdvantage Salmon

Comment Summary: Some comments criticized the EA for the lack of data on sterility of the AquAdvantage Salmon, pointing to studies showing that some triploid females are still capable of producing eggs.

Response: This issue is discussed in the EA in Sections 5.3.2.4 and 7.4.1.3. Although there have been isolated reports of limited gonadal development in triploid fish of several different species, mostly in males, relevant research on triploids of Atlantic salmon and related species indicates functional sterility in females. A very small percentage (0.1%) of triploid Atlantic salmon females may have undergone sexual maturation, but when eggs from these females were fertilized with normal sperm, most underwent little obvious development and none of the fish that developed survived beyond hatching. This confirms that triploid Atlantic salmon females are functionally sterile and do not produce viable offspring. Similar results have been reported from a study on Arctic char (Salvelinus alpinus), a salmonid species related to Atlantic salmon. Therefore, based on the available evidence, FDA has concluded that triploidy would ensure functional sterility and reproductive incompetence in the sponsor’s all-female populations of AquAdvantage Salmon. Further, as previously discussed, all forms of containment should be considered in the context of a containment system in which sterility is a secondary and redundant form of biological containment. The primary form of biocontainment is an all-female population, which effectively precludes reproduction.

(3) Quantitative Failure Analysis

Comment Summary: A few comments indicated that FDA should have performed a quantitative failure analysis on the facilities.

Response: Although FDA did not refer to its analysis as a Failure Mode Analysis in the EA (see discussion of failure mode analysis at III.A.5, above) FDA conducted a risk-based analysis intended to address the same issues by verifying the presence and operational integrity of physical containment systems with two formal inspections of the PEI facility (one in October 2008, and the second in June 2012), as described above. These inspections included review of SOPs and maintenance records to ensure that monitoring and safeguards were in place to guard against failures. Further, as described in detail in the EA, and incorporated into the Canadian Failure Analysis, physical containment at both facilities has redundant containment measures; that is, each component of the containment system has either back-up or complementary features to ensure that escape of GE fish will not occur. For example, as cited above, at the PEI nursery, where eggs are fertilized, triploided, and hatched, and where very small fish are raised, FDA has carefully inspected the physical containment (described in detail in Section 5.3 of the EA) and found that several types of fine mesh screens made of plastic netting or stainless steel are present in the drain lines and fully functional. Chlorine pucks are also used in the drains during the times when fish are spawned.

As described above, Canada’s DFO performed a qualitative Failure Mode Analysis for both the PEI and Panama facilities to determine (1) the potential for acute failure of physical containment; (2) the potential for chronic release of fish; and (3) the likelihood of release while in transit between the Canadian and Panamanian facilities. Based on this analysis, DFO concluded that exposure of AquAdvantage Salmon to the Canadian environment is expected to be negligible with a reasonable certainty. See Section 2.5.1 of the EA. FDA’s risk assessment and the Canadian Failure Mode Analysis followed essentially the same steps, including reviewing the sponsor’s submissions and conducting physical inspections or site visits, and reached the same conclusions. Thus, the Canadian conclusion that the likelihood of exposure of AquAdvantage Salmon to the Canadian environment is expected to be negligible with a reasonable certainty is essentially the same as FDA’s Finding of No Significant Impact.

(4) Potential Theft of Eggs

Comment Summary: One comment noted that even if AquAdvantage Salmon were raised in closed, inland facilities, fish eggs are small, and would be easy to steal by employees or intruders.

Response: In the EA, we state that the two largest risks are malicious release of eggs or fish, or natural disasters. Our inspection of the physical security of the PEI facility indicates that entry onto the premises by unauthorized individuals would be difficult, and penetration into the nursery section of the facility highly unlikely. Even if an unauthorized individual were to gain access, in order for the eggs to survive it would be necessary to remove them in special containers (e.g., coolers) and transfer them to husbandry condition that maintain the tightly controlled conditions for survival; this would be difficult for an individual to accomplish surrepetitiously. Further, there are significant disincentives for employees who would wish to steal the eggs for purposes of release or unauthorized sale, as both would be offenses that would be punishable by civil or criminal penalties.

(5) Enforcement of Containment

Comment Summary: A comment indicated that FDA had failed to take “all possible steps” to ensure containment as required by NASCO, or to provide convincing evidence that it has sufficient powers to apply or enforce conditions on the relevant facilities, or conduct inspections.

Response: The NASCO recommendations on transgenic salmonids are found in the Guidelines for Action Transgenic Salmon contained in the Williamsburg Resolution. The portion of those Guidelines that are relevant to this EA are described above in the Legal, Regulatory, and Administrative Issues section of this comment response, and the relevant record-keeping and reporting are described in more detail the enforcement and inspection components of the approval. See Approval Letter.

(6) Verification and Enforcement of Confinement Measures, Including Ability to Conduct Inspections

Comment Summary: Some comments questioned FDA’s ability to assure and verify that the confinement measures proposed by ABT are continually achieved at the PEI and Panama sites.

Response: As explained in the previous section of this document, ABT must maintain their containment measures as described in their application. Under the FD&C Act, a new animal drug that does not comply with its approved application is “unsafe” and an unsafe new animal drug is “adulterated.” 21 U.S.C. §§ 360b(a); 351(a). The FD&C Act also deems adulterated any food that contains an unsafe new animal drug or a “conversion product thereof.” 21 U.S.C. 342(a)(2)(C)(ii). FDA may take enforcement action against adulterated drugs and foods, including refusing admission to products that appear to be adulterated. If AquAdvantage Salmon did not comply with the conditions under the NADA approval it would be adulterated under these provisions and subject to FDA’s enforcement action.

It is therefore in the company’s best interests to maintain their facilities. FDA’s previous inspections and site visits indicate that the company has adequate procedures in place to ensure the integrity of physical containment, and that its record-keeping has been of equal rigor. In addition, the agency will continue to inspect both of the facilities used to produce or grow-out AquAdvantage Salmon, as well as the Panamanian facility that is used to process the AquAdvantage Salmon into forms suitable for importation into the United States. The PEI facility, also will be inspected by the Canadian government.5 The Panamanian grow-out facility will also be inspected by the Panamanian authorities (EA, Appendix G).

(7) Weather, Climate, and Related comments

i. Effects of Storm Causing “Lost” fish

Comment Summary: Several comments stated that there had been a catastrophic event in Panama as the result of a storm in which many AquAdvantage Salmon were “lost”.

Response: In August 2009, there was a severe storm at the location of the Panama grow-out facility (for detailed information on this event, see the EA, Section 6.1.2.2). During that storm, a tree fell on a pipe that provided oxygenated water to the inland grow-out tanks, and the oxygen level in those tanks fell below the relatively high level required to sustain the AquAdvantage Salmon. As a result, all of the fish at the facility died of asphyxiation; none escaped. There was no failure of any containment systems or equipment. This event was reported in ABT’s public shareholder’s statement (Operations Update dated 15 August 2008) as “all of these fish were ‘lost.’” In this case, “lost” was used to mean “died” as opposed to “misplaced” or “escaped.” As now described in the EA, shortly after the event occurred, ABT implemented a number of corrective measures to ensure there would be no reoccurrence of this event or anything similar. These measures included (1) installation of redundant water supply piping, (2) installation of dissolved oxygen monitors, alarms, and backup oxygenation systems, (3) additional staff training, and (4) development of operational protocols for responding to emergencies and other events. FDA verified the appropriate installation of the new equipment when the agency conducted its site visit in November 2009.

ii. Effect of Weather, Ambient Conditions, or Natural Disasters on Physical Containment

Comment Summary: Other comments indicated that FDA has not taken into full account the variability and severity of weather conditions at both the PEI and Panama facilities, noting that severe weather is common in PEI, and is expected to increase in severity in coming years and opined that the Panamanian facility has recently experienced a physical containment failure due to flooding. Another comment states that the EA fails to consider the seismic activity in the region where the Panamanian facility is located. The comment states that since 2000, Panama has experienced at least 8 seismic events that measured at least 4.5 on the Richter scale.

Response: Climatic conditions and weather-related events, as well as the occurrence of natural disasters, have been described within the EA for both PEI and Panama. See Sections 6.1.1. and 6.1.2 of the EA. The potential for natural disasters to affect physical containment and escape is addressed in Sections 7.2.1.1.4 (for PEI) and 7.2.1.2.3 (for Panama) of the EA.

Although it is possible that the severity of weather-related events, such as floods, may increase in the future, these events are unlikely to affect either facility in such a way as to compromise the integrity of the containment systems because of the locations of these facilities and the redundancy designed into their containment systems. The PEI facility is located approximately 25 feet above sea level and 120 feet inland from the Fortune River, making it highly unlikely that it would be affected by storm surges, tidal waves or flooding events. Similarly, the Panama facility is sited in such a manner (high within its watershed and not in a flood plain) that flooding is unlikely. In fact, there was a significant rainfall and flooding in the immediate area in November 2008 that caused extensive damage at locations further downstream, but contrary to the assertion made by the commenter, the facility itself was not directly affected by flood waters and sustained no serious damage as verified by FDA’s site inspection. No failure of physical containment occurred.

Section 7.2.1.2.3 of EA addresses the issue of volcanic activity. Although volcanic activity cannot be predicted with accuracy, the Panamanian grow-out facility is not located in a high risk area for either pyroclastic flows or lahars. Pyroclastic flows result from the gas and rock emitted from volcanic eruptions, and can exceed temperatures of 1,500°F. Lahars are the result of pyroclastic flows mixing with mud, rock, and water that create a flow of debris that can be meters deeps and flow at several miles per hour. Because both of these phenomena are so destructive, even if they were to impact the facility, it is highly unlikely that any fish would survive due to the very high temperatures and burial resulting from such impacts.

iii. Effect of Water Temperatures on AquAdvantage Salmon

Comment Summary: A number of comments criticized the EA for the lack of the data on AquAdvantage Salmon temperature tolerance. Some of the comments stated that GE fish may have higher tolerance to environmental stressors and survive in the ecosystems that are generally not hospitable to non-GE salmon. Other comments pointed to studies showing that GE coho salmon show altered growth rates at higher temperatures.

Response: The tolerance of AquAdvantage Salmon to physical factors, including water temperature, is discussed in Section 5.2.2.2 of the EA. There is no information to suggest that the temperature tolerance of AquAdvantage Salmon would be greater than that of non-GE Atlantic salmon. In fact, studies have shown that triploid salmonid fishes such as brown, brook, and rainbow trout have a reduced tolerance to chronically elevated temperatures, resulting in a high mortality at temperatures that are not lethal to their diploid siblings. Applied to AquAdvantage Salmon, these results would suggest a lower temperature tolerance for AquAdvantage Salmon because they are also triploid.

(8) Effects of Potential AquAdvantage Salmon Escape

Comment Summary: Several comments indicated that ABT‘s two production facilities are located adjacent to bodies of water that can support GE salmon populations, and that the EA did not consider the effects of possible escapes.

Response: In general, the EA has determined that the likelihood of escape from either the PEI or Panama facilities is highly unlikely due to the multiple physical and, in the case of PEI, physical and chemical barriers. In Section 7 of the EA, FDA discussed at length both the likelihood of escape and the applicable environmental consequences of escape by asking and discussing the answers to four risk questions:

  1. What is the likelihood that AquAdvantage Salmon will escape the conditions of confinement?

    FDA determined that the likelihood of escape would depend on the extent and adequacy of physical containment at each facility. Both FDA and the Canadian government have determined that physical barriers (e.g., stainless steel screens and fine meshes with very small openings) are a critical type of containment in both facilities6. The physical containment methods used in both Canada and Panama are extremely rigorous, durable, effective, and have multiple redundancies to reinforce containment and eliminate the possibility of escape. For example, at the PEI facility, fish are raised in tanks of various sizes within a large building. There is netting on top of each of these tanks and screens on the drains from these tanks to prevent the fish from escaping. In addition, all of the water from these tanks is collected into a single drain line and this water must pass through several stainless steel screens before it is discharged. The openings in these screens are orders of magnitude smaller than the size of a fish, thus making it impossible for the fish to pass through and further preventing them from escaping. Similar screening and barriers to escape are present in parts of the PEI and Panama facilities where eggs are hatched and small fish (fry) are reared. Other containment measures, such as geographical and biological, are useful redundancies to further ensure that AquAdvantage Salmon are prevented from interacting with wild Atlantic salmon. Given the high state of physical containment, its daily monitoring, and the multiple redundancies built into the systems in both Canada and Panama, and for reasons previously discussed in the comments above, we have determined that it is highly unlikely that an escape of any life stage of AquAdvantage Salmon would occur from either facility.
  2. What is the likelihood that AquAdvantage Salmon will survive and disperse if they escape the conditions of confinement?

    The answer to this question is a function of fitness of the AquAdvantage Salmon escaped fish (and life stage) to the specific geographical and geophysical environment into which they would find themselves in the highly unlikely event that they escape physical confinement. Survival and dispersion from the PEI facility is highly unlikely for the earliest life stages (eggs and embryos) due to the relatively high salinity and low temperature conditions in the nearby aquatic environment. Larger fish are not expected to survive if they were to escape due to osmotic shock (they will have lived their entire lives in fresh water and are not expected to survive the sudden, abrupt transition from a low salinity, freshwater environment to a moderately high salinity, brackish water environment), and, based on data from non-GE salmon, would be expected to die off rapidly. Nonetheless, because we have no specific data demonstrating that AquAdvantage Salmon would not survive the change from fresh to salt water, we make the very conservative assumption that they would survive that transition. Nonetheless, there is no evidence that Atlantic salmon are currently present in the local watershed or any nearby watersheds, despite their having been intentionally and periodically stocked there in the past. We have therefore concluded that the current environment in PEI is not highly suitable for the long-term survival and establishment of Atlantic salmon.

    In Panama, the geographical and geophysical conditions in the aquatic environments downstream are inhospitable to AquAdvantage Salmon survival and dispersion. Although the immediate river conditions could be conducive to survival following escape, the temperature of the river in which escaped AquAdvantage Salmon would most likely be found rises substantially further downstream as the river drops in elevation, merges with other rivers, and flows into the Pacific Ocean. In the lower reaches of the watershed, the temperatures are sufficiently high to either limit survival directly, or high enough that feeding stops and the fish would perish from starvation. The watershed ultimately empties into the Pacific Ocean where survival along the Panamanian coast would be virtually impossible as the year-round temperature is at the upper limit for salmon survival. We therefore conclude that the geographical and geophysical conditions present in the aquatic environments surrounding both the PEI broodstock and the Panama grow-out facilities are sufficiently inhospitable to limit the potential establishment and spread of AquAdvantage Salmon to other locations. This is particularly true for the earliest life stages (eggs and embryos) in PEI, which would be unlikely to survive if exposed to high salinity and low temperature conditions in the nearby aquatic environment, and for all life stages of these salmon in Panama, which would be unlikely to survive the high temperature conditions in the lower reaches of the watershed.
  3. What is the likelihood that AquAdvantage Salmon will reproduce and establish if they escape the conditions of confinement?

    In the extremely unlikely event that sexually mature diploid AquAdvantage Salmon escaped from the breeding facility in PEI, and somehow managed to survive the change from the fresh waters in the tanks to the saltier waters outside the facility, it is highly unlikely that they would establish and/or reproduce. We base this conclusion on our observation that no Atlantic salmon (conspecifics), or closely related fish species have been observed in the immediate and many nearby watersheds in many years and sometimes many decades (see answer to Risk Question 2, EA, Section 7.3 and Risk Question 3, EA, Section 7.4).

    In Panama, the potential for establishment is not possible given that the fish are all females, triploid, and cannot reproduce. In fact, no populations of Pacific salmon species (e.g., Chinook, sockeye, or pink) have been observed within several thousand miles of the equatorial Pacific waters into which the nearest rivers flow. With respect to reproduction, however, even in the extremely unlikely event that escaped fish could survive the inhospitable lower river conditions (see EA Section 7.3.1.2), ABT has introduced a level of bioconfinement to guard against reproduction. The EA describes in great detail (Section 5.3) that the fish to be used for production are to be an all-female, triploid population. Triploidy is used as a means to make the AquAdvantage Salmon functionally reproductively sterile, and an all-female population ensures that even if some of the AquAdvantage Salmon eyed-eggs are not triploid (see discussion of triploidy success), they would have no “breeding partners” with which to reproduce. Thus, we have concluded that given all of the available information, there is a negligibly small likelihood that AquAdvantage Salmon would reproduce and establish if they escaped from facilities either in PEI or Panama. Although slightly greater, the hypothetical risk that reproductively competent AquAdvantage broodstock could establish self-sustaining populations is still considered very low given the current lack of Atlantic salmon in the Fortune River and surrounding watersheds.
  4. What are the likely consequence to, or effects on, the environment of the United States should AquAdvantage Salmon escape the conditions of confinement?

    The EA reprises the preceding three risk questions to address Question 4 in Section 7.5, and determines that escape, survival, establishment, and reproduction and movement from either the breeding facility at PEI or the grow-out facility in Panama form the first steps to the exposure pathway to the United States. Because each of these has a negligibly small likelihood of occurring, and those probabilities would need to be multiplied to estimate the likelihood of an impact on the United States, we have concluded no effects on the environment of the United States are anticipated.

(9) Justification for Statements Regarding Starvation of Escaped Farm-Raised Atlantic Salmon

Comment Summary: One comment noted that statements in the EA related to starvation of farm-raised Atlantic salmon that escaped from ocean net pens, attributing this to a lifetime of consuming pelleted commercial diets and their inability to switch to the capture of wild prey. The commenter noted that the reference cited (Muir, 2004) contained no data to substantiate this statement nor were there any data available for AquAdvantage Salmon to address this issue.

Response: Indeed, no studies with AquAdvantage Salmon have been performed with respect to feeding preference and starvation should they escape into natural environments. The cited manuscript (Muir, 2004; EMBO Reports, 5(7): 654-659) addressed this assertion without providing substantiating study results. The EA has been updated and expanded in the applicable sections to substantiate the statements regarding starvation, and why it is believed to apply to AquAdvantage Salmon, citing the relevant published reports related to feeding preferences of farm-raised Atlantic salmon following simulated escape (see EA, Section 7.3.1.1.3).

(10) Extrapolation from Experimental Studies to Actual Environmental Conditions

Comment Summary: Some comments indicated that the FDA EA was extremely limited, and based on outdated methodologies and approaches, and should have incorporated a number of different analyses, including consideration of a growing body of research on genetic and ecological risks of transgenic fish that indicate a high degree of uncertainty in predicting effects of escape due to the inherent uncertainties in extrapolating from experimental (even semi-natural) systems to actual environmental conditions.

Response: FDA recognizes there is a high degree of uncertainty in predicting effects of escape due to the inherent uncertainties in extrapolating from experimental (even semi-natural) systems to actual environmental conditions. Because to date there have been no approvals for AquAdvantage Salmon, free releases of AquAdvantage Salmon into the environment, or approvals or releases of any other genetically engineered fast growing fish, FDA and other risk assessors and regulators must extrapolate to some extent from studies conducted under laboratory or simulated natural conditions, such as mesocosms or simulated streams, taking into account the uncertainty in these studies. Generating additional data in laboratory studies will not reduce this type of uncertainty; thus the FDA analysis in the EA has focused first and foremost on the adequacy and redundancy of the physical containment systems that would prevent escape.

(11) Shipments

Comment Summary: A few comments stated that described containment measures are not sufficient to guarantee the security of the AquAdvantage Salmon eyed-egg shipments, and opined that the EA emphasized physical containment, but failed to address the weakest point in the chain, shipment of eyed-eggs.

Response: There is a full description of the labeling, packaging, and shipping of AquAdvantage Salmon eyed-eggs from the PEI facility to the Panama facility in Section 5.6 of the EA. Briefly, the approval is limited to the shipment of eyed-eggs, which are the life stage most safely transported. The chain of custody of the triploided eyed-eggs includes a double container (a hard, plastic insulated cooler bound with packing straps and then placed inside a heavy cardboard shipping container). A bilingual (English and Spanish) label bearing the statement “Eggs & Fry” and “Not for Resale” includes the product name and provides information on the product identity, limitations of use, warnings, and handling instructions of immediate importance to the end-user.

The package of eyed-eggs is transported by motor vehicle from the PEI facility to a local airport by ABT staff, where prior arrangements with a freight-forwarder have been made to arrange, manage, and personally monitor the air-freight shipment of the product to Panama (including meeting all permits and customs requirements). Control of the package would revert to ABT employees on the ground at a Panamanian airport, who would transport the package by motor vehicle to the Panamanian grow-out facility.

Under the conditions of the approved application, the sponsor has committed to ensuring that the number of eyed-eggs shipped from the Canadian facility is monitored carefully and reported to the agency, as are the numbers of more mature forms of the fish at both the PEI and Panama facilities (see Post-Approval Oversight in this document and the Letter of Approval).

(12) Post- Approval Oversight

Comment Summary: Several comments indicated concerns about the extent of post-approval oversight that could be applied to the two facilities, and how the agency would ensure that commitments made by the sponsor would be enforced.

Response: ABT will be required to meet a number of post-approval requirements. These include monitoring of animal health, genotypic and phenotypic durability, and containment. (See Letter of Approval http://www.fda.gov/AnimalVeterinary/DevelopmentApprovalProcess/GeneticEngineering/ GeneticallyEngineeredAnimals/ucm466214.htm). The requirements also include post-approval record keeping and reporting, including submission to FDA of regular reports concerning the article’s continued compliance with the requirements of the Act as to safety, and that it has the identity and strength, and meets the quality and purity characteristics that it purports or is represented to possess in the NADA. 21 USC § 351. If the new animal drug does not conform with the conditions established in the approved application, the regulated article (in this case the rDNA construct in the AquAdvantage Salmon lineage) is adulterated and any food bearing residues of the drug is adulterated. 21 USC §§ 351(a)(5), 342(a)(2)(C)(ii). FDA may take enforcement action against such adulterated products.

C. Other Comments on EA

(1) Growth Data on Diploid and Triploid ABT Salmon

Comment Summary: One comment indicated that in the 2010 draft EA that was released for the VMAC meeting, growth rate data were included for both diploid and triploid ABT salmon, but that in the 2012 draft EA released for public comment, growth data were only reported for diploid, but not triploid fish.

Response: Figure 3 in both the 2012 draft EA (Page 29) and final EA (Section 5.2.1.3) summarizes the data initially presented in the 2010 draft EA. In the latter presentations (2012 draft EA and final EA), we have included data for the appropriate comparator group: the non-GE, diploid Atlantic salmon representative of farm-raised Atlantic salmon currently in commercial production.

(2) Ensuring Conditions Under Approved Application for Other Purchasers of AquAdvantage Salmon

Comment Summary: Individual comments indicated that there is no guarantee that others who purchase or obtain AquAdvantage Salmon will take proper precautions to ensure no escapes.

Response: FDA’s approval of the AquAdvantage Salmon NADA does not permit production and grow out of AquAdvantage Salmon in any facilities other than those in Canada and Panama. If other parties outside the United States wish to purchase or obtain any life stage of AquAdvantage Salmon with the intent of raising the fish to sell eventually as food in the United States, a supplemental NADA would be required, as explained elsewhere in this document. (see See III. A. (3) regarding potential grow out in the United States).

(3) Assertions Regarding Escapes of Atlantic Salmon on the West Coast of the United States

Comment Summary: Comments were submitted stating that FDA’s EA made incorrect assertions that there had been no escapes of Atlantic salmon on the West Coast of the United States.

Response: The conditions under the AquAdvantage Salmon approved application do not include facilities in the Pacific Northwest, or other areas in which Pacific salmon are found. We believe these comments refer to a report that describes the escape of Atlantic salmon in the Pacific Northwest, primarily in the waters around British Columbia. These were Atlantic salmon fry reared in freshwater hatcheries and moved to sea cages (net pens) in the ocean to be grown to market size. The escapes that are described in this report are thus from those ocean net pens7 (Volpe 2000). The conditions established in the approveda application specify grow-out only in land-based, highly-contained tanks, and not in ocean net pens. Thus, the observations in this report do not apply to this approval.


1  The NADA is for approval of the α-form of the opAFP-GHc2 recombinant DNA construct at the α-locus in the EO-1α line of Atlantic salmon under the conditions of use specified in the application. For ease of reference, this document refers to the application as being for approval of the AquAdvantage Salmon.

2  The Canadian Science Advisory Secretariat conducted a risk assessment and determined that “exposure of [AquAdvantage Salmon] to the Canadian environment is expected to be negligible with reasonable certainty.” See Summary of the Environmental and Indirect Human Health Risk Assessment of AquAdvantage Salmon, CSAS report summary at 18, http://www.dfo-mpo.gc.ca/csas-sccs/Publications/ScR-RS/2013/2013_023-eng.pdf.disclaimer icon

3 Moreover, should the sponsor wish to raise AquAdvantage Salmon at a U.S. location, not only would it be required to submit a supplemental NADA and EA, but it also would have to meet the requirements of all other applicable Federal, State, local and tribal laws and regulations. For example, facilties in the United States would likely require a National Pollutant Discharge Elimination System permit.The National Pollutant Discharge Elimination System is authorized by the Clean Water Act. The NPDES permit program is administered by authorized states. See http://water.epa.gov/polwaste/npdes/.

4  Price, C. 2008. Failure Modes and Effects Analysis. Encyclopedia of Quantitative Risk Analysis and Assessment II. John Wiley & Sons. DOI. 10.10012/9780470061596.risk0475

5 Since 2011, Environment Canada Enforcement Officers have conducted annual inspections of the AquaBounty PEI facility in order to verify physical containment. No violations of Canadian requirements were observed during these inspections. Environment Canada is the primary inspecting agency for physical containment. In addition, with respect to disease-free status, there is an inspection each time an export certificate is requested. These inspections for disease-free status are done on a transactional basis by the Canadian Food Inspection Agency in conjunction with arrangements that have been established by the governments of Canada and Panama, and are outside the purview of the United States government.

6  A summary of the Canadian decision can be found at http://www.dfo-mpo.gc.ca/csas-sccs/Publications/ScR-RS/2013/2013_023-eng.pdfdisclaimer icon DFO. 2013. Summary of the Environmental and Indirect Human Health Risk Assessment of AquAdvatage Salmon. DFO Can. Sci. Advis. Sec. Sci. Resp. 2013/023.

7 Volpe, J., E. B. Taylor, D. W. Rimmer, and B. W. Glickman (2000) Evidence of natural reproduction of aquaculture-escaped Atlantic salmon in a coastal British Columbia river. Conservation Biology 14: 899-903.

Page Last Updated: 11/19/2015
Note: If you need help accessing information in different file formats, see Instructions for Downloading Viewers and Players.