|
A Regulatory Patchwork -
With Big Holes
MICHAEL HANSEN
The present patchwork system for the review of the products of agricultural biotechnology is completely inadequate. Although three federal agencies - the Food and Drug Administration, the Environmental Protection Agency, and the Department of Agriculture - are currently involved in the "review" of the products of agricultural biotechnology, that review is superficial and full of holes, and cannot be relied on to protect either health or the environment.
Because genetic engineering can increase natural toxin levels in food as well as transfer allergens into new foods, and may create other potential health hazards, one would presume that the FDA would assure the human safety of genetically engineered food. Here's what happened instead. In 1992, the agency issued a proposed rule for the regulation of genetically engineered plants. The rule was released at a press conference by Vice President Quayle, who introduced it as a deregulatory initiative. That rule proposed that genetically engineered plants be treated no differently than traditionally bred plants. In other words, no pre-market notification nor pre-market safety testing would be required. The proposed policy functionally put industry on the honor system as to whether they have done the proper safety assessments and testing, requiring them to make judgment calls that should be done by an agency with no commercial interest in the outcome and which is accountable to the public.
The FDA received almost 4,000 comments on its proposed policy, many of which called for safety testing, and the vast majority of which asked that the agency require labeling of all genetically engineered foods. Those making this request included the attorneys general of eight states as well as the American Association of Retired Persons and the trade association Chefs in America. Although the FDA asked for more comments on labeling, it has never finalized the proposed policy. Over the nearly seven years since, basically using the proposed policy, it has reviewed, in a cursory manner, company assessments of some 36 genetically engineered foods, without requiring labeling of any of them.
We feel that agricultural biotechnologies, which move genetic material between completely unrelated plant species, and even between the plant, animal, and microbe kingdoms in ways that could never occur in nature, are not a simple extension of traditional breeding, as FDA alleges in its proposed rule. These differences could increase the probability that certain problems already associated with traditional breeding (such as increased toxic levels) will occur with genetic engineering. These differences also mean that genetic engineering may cause problems never before seen in new plant varieties, such as the potential for the transfer of antibiotic resistance.
The transfer of toxins has already been shown experimentally. In February, there was an uproar in the United Kingdom when the press reported that experiments on rats that had been fed potatoes genetically engineered with the snowdrop lectin gene, previously thought to be nontoxic to mammals, showed adverse affects on the digestive tract. The transfer of allergens has also been documented. In 1996, researchers at the University of Nebraska confirmed that an allergen from Brazil nuts had been transferred into soybeans.
We feel that for all products of agricultural biotechnology, there should be mandatory pre-market notification to FDA, mandatory FDA pre-market safety review, and mandatory labeling. In 1994, after the considerable protest over its laissez faire regulatory policy, FDA actually drafted a proposed pre-market notification rule and sent it to the Office of Management and Budget for approval. However, following the Republican takeover of Congress after the 1994 election, the proposed rule died.
Genetically engineered food could, and should, be regulated under the Food, Drug, and Cosmetic Act. The FDA could require safety testing either by classifying the genetic material inserted into plants as a food additive or requiring a Generally Regarded as Safe ("GRAS") Affirmation Petition. Finally, the FDCA requires labeling when something is a "material fact" to consumers. We believe genetic engineering is a material fact that consumers want to know about. Last May, a coalition of groups sued FDA to remove all genetically engineered plants from the market until they have been adequately safety tested and labeled.
EPA's regulation of environmental hazards is even less adequate. The only class of genetically engineered organisms that it reviews at all is ones engineered to produce pesticidal substances, such as those producing Bt endotoxin. EPA's review, which should cover potential ecological disturbances, "gene pollution" (the flow of genes to non-engineered species of the same variety), and increased insect resistance to the engineered pesticide, is narrow and superficial.
Furthermore, the proposed policy would exempt even the majority of genetically engineered plant pesticides from any regulation. As EPA stated in 1995, "Under the approach articulated in this policy statement, the agency believes that most plant pesticides would not require regulation under [the Federal Insecticide, Fungicide, and Rodenticide Act]…. Similarly under the [Food, Drug, and Cosmetic Act], the agency believes that most plant-pesticides should be exempt from the requirement of a tolerance."
We disagree with such broad and sweeping exemptions, and the rationales behind them. There are literally thousands of natural pesticides in plants that could potentially affect human health and that could be increased from low to high levels within a plant or moved from one plant to another via genetic engineering. EPA's proposed exemptions could potentially lead to health problems in humans and could create new environmental problems, such as poisoning of beneficial insects.
For plants engineered for other traits, such as herbicide tolerance or disease tolerance, EPA does no environmental review at all. This in spite of the fact that herbicide tolerances can lead to increased herbicide use on many crops, and herbicide-tolerance genes can readily migrate-for example they can flow from canola to wild weedy relatives such as wild mustard. Another experiment with a plant in the mustard family demonstrated that genetic engineering increased the rate at which an herbicide tolerance gene flowed into other plants by 20 times compared to conventional breeding. The potential for the creation of "superweeds" is large.
The USDA's Animal and Plant Health Inspection Service is responsible for evaluating potential environmental impacts of field tests of genetically engineered crops, either through permit or notification. The permit process requires a risk assessment and an environmental impact assessment before a field test, as well as potentially requiring environmental data from the field trials themselves. Under the notification process, a permit is not required; researchers simply notify APHIS that they will be conducting a field test. With notification, neither a risk assessment nor specific environmental impact data needs to be turned in to APHIS.
Beginning in the mid 1980s, APHIS required permits for all field trials. However, a study of the data from field trials found that in the vast majority of cases, important ecological issues weren't even addressed. Indeed, it appears that very few quantitative studies were done either prior to obtaining a permit or during the field test itself. In 1995, APHIS virtually abandoned the permit process by stating that any plant species that is not considered a weed can be engineered and tested under the less stringent notification process. Furthermore, APHIS also greatly weakened the process for "determination of nonregulated status," which is one of the final steps in allowing transgenic crops to be commercially produced and marketed.
We feel that EPA, rather than APHIS, should be charged with looking at the environmental impacts associated with field tests of genetically engineered crops and other organisms as the first step in a complete ecological review. Furthermore, EPA should assess the risks on a case-by-case basis, with experimental data, unless and until soundly designed ecological studies show that engineered crops pose no environmental threat. We also recommend that EPA develop a plan for assessing the ecological risks of commercial-scale releases.
New legislation is needed in order to prevent undesirable ecological effects. Such legislation should require actual field data on the probability of gene flow from engineered crops either to non-engineered crops of the same variety or to wild weedy relatives. Furthermore, EPA should be required to look at the effects of engineered plants on non-target organisms. Research in Switzerland, for example, has shown that green lacewing larvae, a major predator of corn pests, were adversely affected when they ate moth larvae that had fed on Bt-corn compared to those fed on nontransgenic corn.
Finally, EPA should be mandated to look at the potential ecological effects posed by all genetically engineered organisms, and to require a full health and environmental review for animals, such as transgenic fish, that are not covered at all by the present patchwork of regulations.
Copyright©1999 Environmental Law Institute®, Washington, D.C. Reprinted with permission from The Environ-mental Forum®. A limited license is hereby granted to viewers to download electronically or mechanically one copy of the work for personal or non-profit educational use, provided that notice of copyright ownership appears on the copy so made. All other rights reserved. Website: www.eli.org
