[To contribute to this conference, send your message to biotech-room1@mailserv.fao.org. For further information on the Electronic Forum on Biotechnology in Food and Agriculture see Forum website.
Note, participants are assumed to be speaking on their own behalf, unless they state otherwise.]

-----Original Message-----
From: Biotech-Mod1
Sent: 21 May 2003 11:18
To: 'biotech-room1@mailserv.fao.org'
Subject: 66: U.S. regulatory experience may be instructive for developing countries

This is from Suzanne Wuerthele, United States.

While Bill Muir is correct (Message 59, May 20) that some products are evaluated "such that a reasonable degree of safety can be determined before they are loosed on the world," the automobile is a poor example. Automobiles were manufactured with no evaluation of future negative ramifications, while prospective benefits were highly advertised. Some countries have wrested incremental corrections at great expense from reluctant manufacturers, thanks to passenger injury and air pollution data. Other societal effects of automobiles, such as poor highway and city design will take centuries to reverse. [Bill Muir wrote "there are a number of ways to estimate risks of potential hazards in a safe secure setting such that a reasonable degree of safety can be determined before the technology is loosed on the world. This is how all new products, including pharmaceuticals, airplanes, cars, baby toys, etc. are evaluated before they hit the market. Biotechnology is no different"...Moderator].

Reinald Doebel pointed out (Message 62, May 20) that actuarial data allows accurate predictions of automobile accident risk, and because we have little experience with recombinant DNA (rDNA) technology, we have no basis for quantitative assessment of its risks. I would add that, like the automotive engineers of the 1920s, we don't yet even know all of the hazards of transgenic organisms.

Developing countries can nevertheless benefit from studying the regulatory experience in other countries. The following thoughts derive from observations in the United States:

1. rDNA technology will likely be used to create inventions which we cannot now imagine, but will want to regulate. For example, pharmaceutical companies are targeting developing countries for growing crops which make drugs and industrial chemicals. Thus, developing countries may want to consider legal frameworks which can regulate any future GMOs. Another example: In 1986 when the U.S. government set up its "Coordinated Framework" to regulate GMOs, it did not anticipate commercialization of transgenic animals. In 2001, when A/F Protein declared its intention to market the transgenic salmon, there still was no regulation which specifically covered that organism. But the Food and Drug Administration (FDA) claimed authority to regulate it by declaring the genes for accelerated growth a "new veterinary drug." Had the transgenic salmon been engineered for some other property, the government might not have had the legal authority to regulate it at all.

2. The U.S. experience also demonstrates the importance of determining in advance who will review the health, environmental, food safety, and social effects of GMOs, and how those reviews will be coordinated. Because coordination among U.S. regulatory agencies is still in flux, it is difficult for U.S. industry and consumers to determine how a product will be regulated or follow it through the regulatory process. For example, it is not clear which agency, if any, will review environmental hazards posed by the transgenic salmon. Likewise, most transgenic plants are reviewed under a law that limits the US Department of Agriculture (USDA) review to the plant's agronomic properties. If the USDA grants a plant "non-regulated status" then there is no legal mechanism for environmental reviews should problems be discovered in the future. And as GMOs go to market, many legal, ethical, and societal issues raised by GMOs are still unresolved. For example, there are no US regulations on liability for the consequences of gene flow to non-transgenic crops.

3. U.S. FDA policy states that GMOs are "substantially equivalent" to their non-engineered counterparts unless they are toxic, allergenic or have significant alterations in nutrient content. "Substantial equivalence" is a legal term. As a result, the transgenes and transgenic products in GM plants are considered "generally regarded as safe" under the law, and do not have to undergo long-term feeding studies or be labeled.

Adoption of a similar "substantial equivalence" policy by a developing country might prevent a food safety review of a GMO that could pose nutritional problems. It would be advantageous for developing countries to define in advance "significant" alteration of nutrients, especially in staple crops comprising a large percent of the diet or in people who have deficiencies. Likewise, developing countries should understand that because the "substantial equivalence" policy precludes long-term feeding tests, the U.S. has no data showing that transgenic foods do or do not create subtle adverse effects when ingested chronically or as a large part of the diet or in people with underlying nutritional problems.

4. Developing countries may want to consider laws which will allow regulation of the whole transgenic plant or animal, not simply its transgenic products. This is illustrated by the U.S. experience with transgenic plants that make pesticides. These plants contain several genes: those directing the pesticide production but also markers and promoters. However, as U.S. pesticide regulations are written, the regulated "pesticide" consists only of the pesticidal chemical and the genes necessary for its production, not other genes or the transgenic plant itself. Thus, if there is a need to regulate the way farmers use the transgenic plant (e.g., create insect refuges), this must be done indirectly. In the U.S., the pesticide manufacturer, rather than the government, inspects and enforces such requirements, then reports back to government regulators. This system was partly to blame for the fact that some farmers who planted StarLink didn't understand that they were required to segregate the corn from the human food supply. Another problem with this system is that it requires multiple reviews of the various genes by separate agencies.

Suzanne Wuerthele, Ph.D., D.A.B.T.
Regional Toxicologist
US Environmental Protection Agency Region 8
Denver, CO
United States
Wuerthele.Suzanne (at) epamail.epa.gov

-----Original Message-----
From: Biotech-Mod1
Sent: 21 May 2003 11:32
To: 'biotech-room1@mailserv.fao.org'
Subject: 67: Re: Methodology for risk assessment

This is from Clark Efaw, United States.

I have some problems with Dr. Blanchfield's (Message 63, May 20) argument in response to Dr. Doebel (Message 62, May 20).

Breeding randomly combines genes that already exist in the species being bred, and that have already gone through countless iterations of checks and balances to get there, both within the organism and in its environment. These checks and balances also decide whether mutations stay or go. The fact that inserted genes are targeted by biotech companies does not have any bearing on their suitability for release into the environment. They are targeted for reasons of economic benefit, and the producers seem to be very deliberate about avoiding any responsibility for economic or environmental consequences that can occur on release (see Julie Newman, Message 61, May 20).

Likewise, the analogy to airplane technology looks weak. The first flights at Kitty Hawk and in the following decades only exposed one human at a time to risk. Now, nearly a century later, most of us still have the option of avoiding it should we so choose, safe or not. Maybe a better analogy would be that of electricity distribution. Our great-grandparents were concerned that if all buildings were supplied with electricity, people would be dropping like flies. Today most homes and businesses in the developed world are flowing with juice, and, yes, it kills tens to hundreds of people every day. A great deal of environmental damage is caused in the production and distribution of elecric energy. Species are driven to extinction because of it. But we have made that trade-off because to a large extent we consider it worth the price. We minimize risk by isolating the hazard and holding product manufacturers responsible for product safety. We allow communities to have a say in how it is produced and distributed. We keep it contained, and label it clearly. We have government agencies dedicated to keeping it safe and to controlling monopolistic practices. We take advantage of the possibilities while going to great lengths to build redundancy into safety measures. These conventions have evolved over more than a century of use, and, yes, they use a different standard in Europe than in the US. I wouldn't take this analogy much farther, but the price of caution in the use of any new technology like biotech should be weighed against the benefit. If it's too expensive to internalize the costs of caution, then it's not yet worth it economically.

Clark Efaw,
CARE
Atlanta,
Georgia
United States
efaw (at) care.org

-----Original Message-----
From: Biotech-Mod1
Sent: 21 May 2003 11:48
To: 'biotech-room1@mailserv.fao.org'
Subject: 68: Substantial Equivalence

This is from C Kameswara Rao, Bangalore, India.

The Principle of Substantial Equivalence (SE) and Genetically Engineered Organisms (GEOs):

The Food and Drug Administration (FDA) of the US used the Principle of SE for decades to assure the public of the safety of foods (and drugs). The stringency of the regulatory oversight and safety standards of the FDA are regarded as high, and most other countries routinely approve drugs and pharmaceuticals on the basis of approval by the FDA. Subsequently, the Principle of SE has been applied to foods and other products from GEOs, to assure the consumer that the product is 'substantially equivalent' to its conventional counterpart and that it is safe for human consumption. This certification refers only to the product and not the process of its production.

The FDA has long considered GE plants to be substantially equivalent to conventional varieties and has published a policy statement to the effect that no other regulatory review to assure the safety of foods from GEOs is deemed necessary. However, taking advantage of the provision for voluntary consultation, biotech companies in the US seek independent certification by FDA of all GEO varieties and their products that are marketed in the US.

The policy of the FDA did not result in any health concerns but invited criticism on account of a) the FDA has a mandatory process for approving transgenic animals, and b) the United States Environment Protection Agency (EPA) and the United States Department of agriculture (USDA) have a mandatory and open process for evaluating the biosecurity of transgenic plants.

Foods from GEOs on the US markets have been tested extensively and judged substantially equivalent to their conventional counterparts. This is the case with many products from GEOs, such as cotton oil, tomato and corn from Bt varieties. Some products may contain miniscule quantities of one or two additional proteins, which are broken down during processing or digestion, or some others may contain some compounds not occurring in the counterparts that are present but below threshold levels. Such products are categorised as 'Generally Recognised As Safe' (GRAS). A factor that would cause a product to be considered otherwise is the presence of genes in the GEO or its product, which would code for fats, proteins or carbohydrates that may be toxic or may cause allergies or may change the nutritional value of the product. Bt potato is one such where the gene expresses in the tuber, but this has been tested and considered safe. For these reasons, a product can be certified as substantially equivalent to its counterpart only after an extensive analytical study.

Certifying a product as SE or GRAS is to assure the consumer that the food is safe to consume. While in the US, on account of the policy of the FDA, no labelling as SE or GRAS is mandatory, it is not so in several other parts of the world. This dichotomy causes considerable confusion in the global policy, and leads to needless controversies. There is a dire need for international harmonisation and uniform policy.

In recent times suggestions were made for the application of the Principle of SE to all products of genetic engineering, including livestock feed and GE crops, which raises certain questions.

When the principle of SE is applied to a GE crop variety, it should be substantially equivalent to its isogenic variety, in genotype, marked characters and performance, but for the transgenes and their anticipated characteristics and benefits. [Isogenic lines are genetically nearly identical, except with respect to identified genes...Moderator]. The objective is that, if the isogenic was safe, the transgenic would be equally safe, provided that the newly introduced transgenes do not exercise any adverse effects by themselves or through affecting the expression of any other genes of the isogenic counterpart. Such an assurance requires scientific evaluation of the crop variety first, and then of its products. This involves additional efforts, time and expense, raising costs to the consumer.

All US agricultural biotechnology companies submit to the FDA, voluminous dossiers on the safety and risk analysis of the GEOs and their derivatives produced by them before they go on the US markets. Such a voluntary mechanism should be global. However, like drugs and pharmaceuticals, what is considered safe in one country should be so in other countries, provided uniform testing procedures are adopted. There is no need to repeat every test in every country.

Another consideration is that transgenics would be substantially equivalent to their isogenics, up to a point in real time. Mutations occur naturally and randomly, affecting all genes. Members of the same population would be subjected to mutation of different genes and so would not be entirely identical with each other. Lethal mutations are naturally eliminated. If any genes related to the desired characteristics mutate, they are eliminated in the process of selection, but those that do not affect the desired characteristics escape attention and accumulate. After a certain number of generations, SE cannot be maintained for the entire genotype of all members of population. A critical genetic analysis at that point of time will contravene SE, although SE can be established for the genes of the desired characteristics. Such a situation would cause problems in some countries, where the regulatory authorities apply the principle in letter and a lot more strictly than in other countries.

On account of these reasons there is a need to re-examine the issue of SE and for re-defining its applicability to GE crop plants and their products, laying emphasis on a reasonable application of the principle, addressing only those genes and their products that are relevant to the objectives of developing a particular transgenic variety or product.

Since the objective of technology deployment is improvement of a product over what exists, the question should be 'is this better and safer than what is already there?'

Professor C Kameswara Rao
Executive Secretary
Foundation for Biotechnology Awareness and Education
Bangalore 560 004,
India
krao (at) vsnl.com

-----Original Message-----
From: Biotech-Mod1
Sent: 21 May 2003 14:28
To: 'biotech-room1@mailserv.fao.org'
Subject: 69: Re: Methodology for risk assessment

This is Dr. Doebel replying to Prof. Ralph Blanchfield's response (Message 63, May 20) to my earlier argument (Message 62, May 20).

I apologize to the participants for insisting on some seemingly minor details not directly related to regulation - but as Prof. Blanchfield clearly demonstrates with his repetition of an earlier message, nr. 58, May 19, (that time in response to Dick Richardson's message (nr. 51, May 13) on the molecular mechanics of genetic engineering versus the molecular mechanics of "traditional breeding"), it is the interpretation of such details which furnishes arguments for or against certain types of regulation. In particular: whether regulation should mean HOW to introduce a certain GMO or whether it should also be concerned with WHETHER a certain GMO should be introduced at all.

Prof. Blanchfield rejects the statement that: " 'breeding' leaves the chemical processes of adding a new piece of information to the interaction between the organism and the environment". He points to irradiation methods and polyploidisation as "conventional breeding" and writes "Instead of inserting specific known genes in a targeted way, "breeding" inserts an unknown collection of genes in an untargeted way. Whatever the problems of genetic modification, they are at least matched if not surpassed by those of "breeding"."

The scientific facts - as I see them - are that:

1) Irradiation methods are not part of "conventional breeding" but a rather recent introduction of forced mutation - leaving the response to the environmental (artificial) stimulation through irradiation to the DNA: not all mutations will survive and become part of the "real" breeding programmes which attempt to multiply those mutations producing particularly desirable traits.

2) Polyploidisation refers to stimulating the cell to produce an additional set of identical molecules of DNA - and therefore does not touch the internal structure of the DNA molecules. These are also rather recent methods and do not refer to "breeding" in the traditional sense of selecting plants with desired traits over plants with less desired traits.

3) The claim that the new methods consist of "inserting specific known genes in a TARGETED way" is simply not true: the truth is that snippets of the DNA molecule from one species (which are believed to result in a specific desired expression/action/trait of the whole organism - such as producing a chemical toxic for insects in the case of Bt cotton) are introduced into the sequence of nucleotides of the receiving organism (from another species) in a NON-TARGETED way. Non-targeted because (unless a new - and as yet little publicised - method has been found in the past two years) the specific location of insertion of the "alien" piece into the sequence of the receiving DNA cannot be determined in advance. This is problematic in an UNPRECEDENTED way, because the overall organismic reaction is determined not only by the chemical composition of the inserted snippet, but also by the interaction of this snippet with the rest of the molecule: the sequence of nucleotides determines the SHAPE of that molecule, and the shape in turn determines both the interaction within the molecule and the interaction between the molecule and the environment.

I do believe that arguments for more "enabling" regulations should respect known scientific facts, rather than referring to possible future benefits of the technology in question.

Dr. Reinald Doebel
Institut fuer Soziologie
Scharnhorststr. 121
48151 Muenster
Germany
personal e-mail: dobel (at) uni-muenster.de

-----Original Message-----
From: Biotech-Mod1
Sent: 21 May 2003 16:57
To: 'biotech-room1@mailserv.fao.org'
Subject: 70: GM regulations based on probability of spread

Professor Muir again.

The statement by Reinald Doebel (Message 62, May 20) "the probability for the occurrence of a 'risk' must be known. And it can be known only on the basis of what has really happened in the past" and reflected in other messages, implies that the regulatory process for GM products should be different from that of conventional breeding because the risk of harm (environmental) is unknowable for GM products because there is no past experience and cannot be determined until after the fact.

I differ with that conclusion on the basis that the longest running experiment in existence is evolution as a result of natural selection. We have studied the process of natural selection for decades and know how it works. Based on that knowledge we can develop regulations which restrict or prohibit release of GM organisms that appear to have some fitness advantage in a natural setting, e.g. forest trees that produce their own insecticide. It is clear that such a GM plant would have a survival advantage, it can be measured. Thus, regulations can be based on probability of spread of GM plants or animals into ecosystems. I feel we can easily quantify this risk of spread into categories of high versus low. Shifting the regulatory process to one that acknowledges the evolutionary process in an ecological context would help alleviate society concerns. This regulatory review could focus on the product and not the process, i.e. both domestically bred and GM plants and animals could be put through the same set of tests.

Bill Muir, Ph.D.
Professor of Genetics
Department of Animal Sciences
Lilly Hall
915 W. State Street
Purdue University
W. Lafayette IN 47907-2054
United States
http://icdweb.cc.purdue.edu/~bmuir/
bmuir (at) purdue.edu