Critique of US Biosafety Regulation and Implementation

Critique of US Biosafety Regulation and Implementation

Doreen Stabinsky, PhD

Greenpeace International

College of the Atlantic, USA

Introduction

The United States government consistently promotes its regulatory framework for genetically engineered organisms as comprehensive and strict. Government and industry claim that environmental harm from transgenic organisms is adequately prevented by thorough risk analysis on the part of the regulatory agencies charged with this work. However, a careful review of the US regulatory framework leads to other conclusions. In this paper we detail the shortcomings and loopholes of the US regulatory system.

To analyze the US system, or in fact any national regulatory system, a number of criteria defined at the international level can serve as a framework for analysis. We can evaluate the US regulatory system by asking how well, or if, the system incorporates generally recognized components for preventing harm from the introduction of GMOs into the environment. We suggest three criteria to use here that derive directly from the Cartagena Protocol on Biosafety, the international agreement on transboundary movement of engineered organisms:

•Risk assessment – identify and evaluate the potential adverse effects of living modified organisms on the conservation and sustainable use of biological diversity in the likely potential receiving environment, taking also into account risks to human health. (Article 15, Cartagena Protocol)

•Risk management – establish and maintain appropriate mechanisms, measures and strategies to regulate, manage and control risks … associated with the use, handling and transboundary movement of living modified organisms. (Article 16, Cartagena Protocol)

•Precaution – many countries and international bodies have also adopted the precautionary principle as an overarching principle for addressing the risks of GMOs. There are numerous legal formulations of the principle – the basic idea, as articulated in Article 11.8 of the Cartagena Protocol, is that the:

Lack of scientific certainty due to insufficient relevant scientific information and knowledge regarding the extent of the potential adverse effects of a living modified organism on the conservation and sustainable use of biological diversity in the Party of import … shall not prevent that Party from taking a decision, as appropriate, … in order to avoid or minimize such potential adverse effects.

The US Coordinated Framework and the implementing agencies

In 1986 the US government developed a framework for regulating GMOs that they called the Coordinated Framework. It was an elaborate patchwork quilt of already existing laws that they would use to regulate GMOs instead of making new laws. Some of the laws in that crazy patchwork quilt dated from as early as 1930. As one could imagine, there were and continue to be difficulties in applying old statutes, written for completely different purposes, to the regulation of completely novel organisms that pose novel risks. Some of these problems are highlighted below. There are three main agencies involved in regulating GE food: the Environmental Protection Agency (EPA), the United States Department of Agriculture (USDA), and the US Food and Drug Administration (FDA).

EPA

The Environmental Protection Agency regulates GMOs under two main statutes: the Federal Insecticide, Fungicide and Rodenticide Act (FIFRA) and the Toxics Substances Control Act (TSCA). Under FIFRA, the EPA has regulatory authority over pesticidal plants – that is, plants that produce their own pesticides, such as Bt corn and cotton. In fact, the EPA regulates pesticidal plants in the same way it regulates chemical pesticides. The USDA has regulatory authority over all other transgenic plants.

USDA

All transgenic plants were originally regulated under a law called the Plant Pest Act of 1930. The law has recently been revised and is now known as the Plant Protection Act.

From the start, the USDA has been limited to regulating GMOs as “plant pests.” This is not a particularly useful limitation from a biosafety perspective, as the range of potential biosafety impacts that a GMO might have is much broader than what a “plant pest” might cause. For example, impacts on non-target organisms like butterflies are not usually considered as impacts of the pest of a plant.

The Plant Protection Act does not provide scope for regulatory oversight at the commercialization stage. Therefore, at some point in the regulatory review, the government has to decide that the GMO is not a plant pest. At that stage, when the crop is ready for commercialization, it is exempted from further regulation.

In terms of biosafety, however, it would be appropriate to continue monitoring for potential impacts after the crop is commercialized. There is no reason to think that small-scale field trials would provide information about effects that might occur during large scale planting. But to require monitoring and submission of data to regulatory agencies, you need to still have regulatory control over the GMO. In the US, there is none – at the commercialization stage, GMOs are considered “non-regulated articles.” The risk management strategy for the United States is something like “if we see a problem we will do something about it,” but there is no systematic monitoring for effects.

However, the USDA has declared that crops producing pharmaceuticals and industrial chemicals will not be deregulated. Even when the product is commercialized (right now several chemicals derived from pharmaceutical plants are commercially available, but none of them are approved for human drug use), the USDA will retain regulatory authority – no exemption will be considered.

FDA

The principal statute used by the FDA to regulate genetically engineered organisms is the Federal Food, Drug and Cosmetic Act (FFDCA). However, there is little formal regulation of genetically engineered foods by this agency. There is no pre-market testing required for engineered foods, no requirement for labeling, and no requirement for FDA approval before an engineered food is placed on the market.

Although the FDA does little to nothing to regulate GMOs before they enter the food supply, they have an interesting role to play with respect to the biosafety considerations of genetically engineered animals. The Bush Administration has decided that the FDA will be the agency to regulate the ecological and human health risk assessment of genetically engineered fish.

The FDA and fish. If you are using old statutes, you need laws in place that already somehow regulate the organism in question. There are no laws in the United States governing the release of fish into the environment, so when the government was faced with an application for release into the environment and onto the market of genetically engineered salmon, US regulators scrambled to figure out a way to regulate the salmon. The FDA decided they would use the animal drug law, as the salmon produce growth hormone – that is, they will regulate the salmon as if it were an animal drug. That is how the Food and Drug Administration came to be the regulatory body deciding whether GE salmon pose a biosafety risk.

Major environmental risks considered by US regulatory authorities

As might be imagined from the preceding discussion about the limitations of statutory authority on the part of the main US regulatory authorities, there are also limitations as to how potential biosafety problems are defined. The major categories for regulatory concern are a small subset of those concerns identified by scientists and policymakers at the international level and include:

• Weediness
• Pest resistance
• Impacts on non-target organisms
• Gene flow
• New weeds

We provide more detail below about how the agencies address these environmental risks. We follow this section with a discussion of critiques of the EPA and USDA by major scientific bodies.

USDA

The USDA has focused most of its regulatory concern on whether a transgenic plant might develop into a weed. Because of their statutory obligation to focus on plant pests, this is not surprising. Unfortunately statutory obligations are not generally the main reason for regulatory consideration of any particular risk by national or international authorities, but rather the overall biosafety hazard a crop might pose.

USDA risk assessment has historically been based on literature review and comparison with characteristics of weeds (does it look like a weed), rather than based on results of actual experiments. This armchair analysis has met with significant scientific criticism of the agency’s cursory reviews starting in the early 1990s. The lack of scientifically sophisticated analysis has left the agency wide open to criticism. Moreover, actual experimental work now suggests that there are increases in fitness of various transgenic crops, countering earlier agency armchair assertions that any inserted gene was certain to have deleterious effects on fitness.

The USDA has paid no attention at all to the potential for development of

new weeds due to the development of resistance to herbicides used with herbicide-tolerant crops. A quick perusal of agronomy and farm production magazines, of course, provides too many examples of the many new weeds currently developing in herbicide-tolerant production systems in the United States.

EPA

In its risk assessments of engineered organisms, the EPA has primarily addressed the questions of the evolution of pests resistant to insect-resistant plants and the potential impacts of pesticidal plants on non-target organisms. In general, the EPA has been relatively more concerned about the environmental impacts of GMOs than the USDA, but not to the point of stopping the commercial introduction of any transgenic plant variety.

The EPA has recognized the need for regulatory measures to delay the onset of pest resistance, specifically the requirement that farmers plant refuges to harbor populations of susceptible insects. This policy direction comes in response to heavy lobbying efforts by environmental and organic farm organizations. The EPA has not, however, placed significant emphasis on implementation of the refuge requirements, nor have they followed advice of their scientific advisory panel on the optimum size of refuges. Compliance with refuge requirements among corn farmers is far from complete, with recent surveys suggesting that only 80% of farmers are planting appropriate non-GMO fields. Scientific advisors suggested up to 20% of cotton fields should be planted to non-Bt cotton; the EPA has required farmers to plant 5% non-GE.

With the publication in Science magazine of an article demonstrating impacts of Bt corn pollen on Monarch butterflies, the EPA was forced to pay attention to non-target impacts. The regulatory agencies commissioned several field studies to investigate the potential scope of impacts. The field studies showed that Bt corn pollen from several varieties was not present in actual production situations at a level that would be harmful to the butterflies. However, the studies also showed that one variety, Bt176, expressed such high levels of Bt in the pollen that indeed it would be toxic to butterflies at pollen concentrations normally found in the wild. The EPA did not immediately remove Bt176 from the market, but allowed the company to let its permit expire and gave it several years to get rid of existing seed stocks.

There has been consistent scientific criticism of the agency regarding its industry bias and reliance on scientifically questionable industry studies that do not have adequate statistical power to determine effects. Impacts on non-target organisms that have been demonstrated experimentally, and published in peer-reviewed journals have been ignored. Some of these scientific criticisms are discussed later in this paper.

USDA and EPA on gene flow

Interestingly, both USDA and EPA have shown some concern over the problem of gene flow from transgenic plants to their wild or weedy relatives. For the two dominant crops of maize and soy, this is not a conflict with their promotional approach to genetically engineered crops, as there are no wild relatives of these crops in the United States. However, the development of maize varieties engineered to produce drugs and industrial chemicals has forced them to look at the issue and the USDA has taken a slightly more cautious approach to these types of GE crops.

In the case of crops where wild and weedy relatives exist, apart from canola, the agencies have been much more cautious. There has been no movement in the regulatory system of transgenic sunflower, although at least one company – DuPont/Pioneer – has an ongoing research effort on this crop. There is a clear potential for gene flow to wild relatives of sunflower in the United States, and there have been published studies showing fitness increases in wild sunflower that receive an insect-resistance transgene.

Cotton is another special case, one where the EPA has actually taken precautionary measures to prevent gene flow to wild relatives (in Hawaii) and weedy relatives (in Florida, Puerto Rico and the US Virgin Islands). The EPA has placed significant restrictions on the cultivation of Bt cotton (Bollgard), stemming from a regulatory review during the recent re-registration of all the Bt crops. In the words of the agency:

“adequate data do not exist to complete a full risk assessment on the effects of the Bt cry1 Ac protein in wild cotton. Until thorough research on the impacts of gene flow can be completed, restrictions on where Bt cotton can be planted are being implemented.

In light of the lack of basic biological data (e.g., pollinator ecology, compatibility/sterility factors, potential impact of Bt on herbivores, distribution of native populations) on G. tomentosum, the wild Hawaiian cotton, conservative measures are needed to mitigate hybridization with cultivated cotton on these islands. Similarly, the paucity of data on the distribution of feral cotton in the US Virgin Islands and Puerto Rico indicates the following terms and conditions must be instituted to mitigate gene flow concerns:

no planting of Bt-cotton south of Route 60 in Florida,

commercial culture of Bt-cotton is prohibited in the state of Hawaii,

test plots or breeding nurseries established in Hawaii must be surrounded by 24 border rows of a suitable pollinator trap crop regardless of the plot size and must not be planted within 3 miles of Gossypium tomentosum,

commercial culture, experimental plots and breeding nurseries of Bt-cotton are prohibited in the US Virgin Islands, and

commercial culture of Bollgard cotton is prohibited in Puerto Rico. Test plots or breeding nurseries established on the island of Puerto Rico must be surrounded by 24 border rows of a suitable pollinator trap crop regardless of the plot size and must not be planted within 3 miles of feral cotton plants.”

The problem of maize

In 2001, the world was shocked to find out that traditional varieties of maize in Mexico, in a center of origin and diversity of the crop, were contaminated with one or more transgenes. The source of the transgenes was maize imported into Mexico from the United States.

While there is no outcrossing danger in maize in the United States (except for the current problem of pharmaceutical maize), the cross-border impacts are huge. However, although the EPA has taken extraordinary precautionary measures in banning the cultivation of Bt cotton in Hawaii, where an endangered species of cotton lives, the agency has done nothing to protect the center of origin of one of the world’s most important food crops. The double standard of protecting cotton and leaving maize completely unprotected is conveniently explained away by the fact that Mexico is outside of the area of the world that the agencies are legally obligated to protect. The incalculable worth to the world of maize diversity is not taken into consideration at all in regulatory decisions. Attention has been called to this gaping hole in rules to protect valuable crop diversity by none less than the Commission for Environmental Cooperation of the North American Free Trade Agreement (NAFTA):

Information and knowledge on the attributes and risks of any new crop cultivated in all three countries is needed before such a crop is commercialized. … To ensure complete regulatory oversight, there should be greater information exchange among regulators in the three countries in order that no products are released without the knowledge of all three governments. Ideally, harmonization should address risks both specific to individual countries and those common to one or more of the countries.

Scientific critiques of US risk assessment and risk management

As noted several times already, there have been consistent critiques of US regulation of GMOs from within the scientific community in the United States. We provide examples from a few of these critiques below:

NationalAcademy of Sciences and the USDA

The United States’ National Academy of Sciences has undertaken several evaluations of the US system to regulate genetically engineered organisms. Each study has been critical of the lack of scientific foundation of regulatory decisions. For example, the NationalAcademyreport, “Environmental Effects of Transgenic Plants: The Scope and Adequacy of Regulation” takes the USDA to task for regulations that lack scientific justification and are not applied uniformly. Regarding plants genetically engineered to be insect resistant, the academy found that USDA’s assessment of certain environmental risks was “generally superficial.” The academy’s scientists also found that USDA’s process “hampers external review and transparency” by keeping confidential too much information on environmental assessments in order to protect corporate trade secrets.