Long Term Monitoring of Health Effects Related to Genetically Modified Foods in Canada

Post Market Surveillance and the Role of Labeling

A report prepared at the request of the Canadian Biotechnology Advisory Committee

Leonard Ritter, PhD
Executive Director
Canadian Network of Toxicology Centres
and
Professor and Associate Chair
Dept of Environmental Biology
University of Guelph

Top

Executive Summary

The topic of genetically engineered foods continues to attract broad attention from government policy makers, the scientific community and the general public. To many, the heart of the debate centers on the benefit, or potential harm, that such food may present to the human population. It is therefore not surprising that the methodology to evaluate the safety of such foods, the value of post market monitoring programs and the role of labeling have become the central focus of much of the public debate.

Regulatory scientists, learned societies and prominent scientific authorities all agree that the current toxicological paradigm is inadequate to assess the human safety of genetically modified whole foods. Similarly, there is also broad consensus that potential allergenicity, the primary issue of concern regarding human exposure to genetically modified foods, cannot be reliably evaluated utilizing currently available laboratory animal models. Epidemiology, a valuable tool for the study of cause and effect for a broad range of exposures and adverse health outcomes, lacks sufficient sensitivity to distinguish effects uniquely attributable to genetically modified foods against a background of adverse health effects already known to be associated with lifelong exposures to conventional diets.

The role of labeling in the development and marketing of genetically modified foods has received wide attention as well. Mandatory labeling of genetically modified foods - where practical and possible - may be useful in mitigating against adverse health effects, but only in those instances where there are clear and scientifically established health risks posed by the novel food product. In general, labeling should be viewed as an issue of "informed choice" or a means of engaging the general public in the risk management process, rather than as an alternate for inadequate scientific methodology that precludes appropriate pre-market safety assessment of genetically modified foods.

In conclusion, the methodological tools available to monitor and detect long-term adverse health effects are limited and crude, and likely to be of only marginal value. Rather, it may be more productive to focus research efforts on the development of methods that can reliably detect the impact of unintended effects, and to make these unintended effects the principle focus of long term post market monitoring programs.

Top

Long Term Monitoring of Health Effects Related to Genetically Modified Foods in Canada

Introduction

The science of toxicology seeks to identify potential human health hazards, and quantify the risks, that may result from exposure to a broad array of anthropogenic and synthetic toxicants, including novel foods, and novel constituents of traditional foods, resulting from the application of the principles of biotechnology to foods of plant origin. Although biotechnology has been used for centuries, through the application of selective breeding and other cultural practices, direct genetic manipulation of plant cells probably traces its origins to a recent history of only about twenty-five to thirty years.

Regulatory toxicology describes a series of studies which, when carried out prior to market introduction of a new product, seek to predict the likely behavior and disposition of new and novel chemicals, and complex mixtures of chemicals, in the human body. The principle objective of these studies is to define and characterize the potential for natural and synthetic substances to cause adverse health effects and to assess and characterize the biological plausibility and level of risk that humans might experience under a specified set of circumstances. These studies have evolved over the last thirty years to now include endpoints that describe effects ranging from toxicity following acute exposures to single high doses of the toxicant, studies that investigate the likelihood that a substance may induce irritancy (to eye or skin) or an allergic response in a vulnerable sub set of the population, and studies designed to describe the potential for a chemical to induce short term, chronic toxicity or carcinogenicity. In order to overcome statistical limitations of laboratory experimental study design, toxicology studies are typically carried out under carefully controlled conditions that include high dose exposure to small groups of laboratory animals. Current approaches to safety assessment of chemicals have developed primarily for the evaluation of the toxicity of single chemical entities. While the classical toxicological paradigm may be adequate for the safety assessment of individual novel constituents of foods that may result from genetic manipulation, whole food is a complex mixture of many constituents. Noting that the toxicological paradigm may not be applicable to the safety assessment of whole foods, several authors (FAO-WHO, 2000, RSC, 2001; US SOT, 2002) have concluded that conventional toxicological studies are of little value in assessing the safety of whole foods, including both traditional and genetically modified foods as well. Moreover, it is widely recognized that one of the pre-eminent safety concerns about food derived from genetically modified crops is the issue of potential allergencity (US SOT, 2002) and the inadequacy of the current toxicological approach to reliably predict potential adverse health outcomes resulting from exposure to genetically modified foods on the basis of animal models currently available to the scientific community (RSC, 2001; US SOT, 2002; FAO-WHO, 2001).

Top

The scientific method is inherently fraught with uncertainty. As noted above, the toxicological paradigm cannot fully address or predict important health issues, and even greater uncertainty clouds our ability to predict allergenicity. Long term monitoring to identify important potential adverse health effects that cannot be reliably predicted within the context of the uncertainties inherent in predictive toxicology testing is a well-established component of public policy. Health Canada recognizes the limitations of the scientific method on which its safety evaluation of therapeutic products is built, requiring mandatory reporting of all adverse drug reactions observed following large scale human exposure to fully approved therapeutic agents, to the Canadian Adverse Drug Reaction Monitoring Programme, which has collected information on adverse drug reactions since 1965 (Health Canada, Adverse Drug Reactions Bulletin, 2000) Similarly, in recently announcing a new legislative and regulatory agenda for the management of pesticides in Canada, the Minister of Health noted the need to strengthen post market control of pesticides by requiring that adverse effects be monitored and reported after a pesticide product has been reviewed and registered for use in Canada by the designated authority (Health Canada, News Release, March 21, 2002).

This report will examine the implications of the inadequacy of the toxicological model for the assessment of potential adverse human health effects in the context of long term health effects monitoring. This report will also review the role of labeling, if any, as a component of long-term health effects monitoring associated with human exposure to genetically modified foods of plant origin. Finally, this report will provide a brief commentary on the application of the precautionary principle to genetically modified foods.

Top

Long term monitoring – non allergenicity effects

In March 2000, the Joint FAO/WHO Food Standard Programme of the Codex Alimentarius Commission organized an ad hoc intergovernmental task force on foods derived from biotechnology (Codex Alimentarius Commission, 2000). The Task Force noted the need to improve animal test methods for safety assessment of complex new food products that would result from the application of biotechnology to foods of plant origin. The Task Force also referred several other issues on the safety aspects of genetically modified foods to a joint FAO/WHO expert consultation, which was convened in late spring, 2000.Amoung other issues, the expert consultation considered what scientific approach could be implemented to monitor and assess possible long-term health effects. In considering this question, the expert consultation noted that little was known about long-term effects of any foods, including, but not restricted to, the long-term effects of consumption of foods derived from biotechnology. Much debate has prevailed in the scientific literature as to the ability of the conventional toxicological paradigm, and the current state of knowledge, to provide reasonable assurance that conventional or genetically modified foods are indeed safe. The consultation expressed the view that it was unlikely, on the basis of current knowledge, that long term effects could be specifically attributed to genetically modified foods. The consultation also expressed the view that observational epidemiological studies would be unlikely to identify any adverse health effect against a background of undesirable health effects already associated with exposure to conventional foods. Observational epidemiological studies, the usual approach for the study of disease distribution in a population, only allows the investigator to observe the occurrence of an adverse effect in a population who are already segregated into groups on the basis of some unique exposure characteristic (smoking vs. non smoking). The difficulty with this study design, as recognized by the expert consultation, is that the observed group in this study design usually differs in some characteristic in addition to the specific factor under study. As many of these additional factors, such as exposure to genetically modified foods, are often not measurable, and frequently not even known, the role of a specific factor on disease outcome is very difficult to demonstrate (Mausner and Kramer, 1985). Other post market monitoring study designs, such as randomized controlled trials could be used, if properly designed and conducted, to gather post market information on the medium and long term effects of any foods, including effects resulting from exposure to genetically modified foods. While such studies might provide additional useful information on which to assess post market safety of genetically engineered foods, randomized controlled trials can be very costly, are very difficult to conduct and would, in any case, be compromised by competing variables influencing health outcomes, such as variation in diets and dietary components, on an ongoing and long term basis. These limitations to the use of epidemiological studies to monitor health effects are further exacerbated by wide genetic variability, and susceptibility, in the general population thereby creating lesser or greater predisposition to food related effects, regardless of the source of the food.

Top

Given the prominent role of diet as a significant etiologic risk factor for serious diseases (Canadian Cancer Society, 2001) there can be little doubt of the merit of post market monitoring to investigate the relationship between diet and disease for all diets, including those comprised of conventional and genetically modified foods. It is, however, equally important to recognize that many, indeed most, serious chronic diseases are multi-factorial in etiology. While post market monitoring studies could, at least in theory, be useful for identifying adverse effects in particularly sensitive sub sets of the population (Amanor-Boadu, 2002) that could not be detected in pre market laboratory studies, observational epidemiological studies would likely lack the sensitivity to isolate that component of disease to which exposure to genetically modified foods could be uniquely related, especially against a background of adverse effects associated with exposure to conventional diets.

It will be clear that the methodological tools available to monitor and detect long-term adverse health effects are limited and crude, and likely to be of only limited value. Rather, it may be more productive to focus research efforts on the development of methods that can reliably detect the impact of unintended effects, and to make these unintended effects the principle focus of long term post market monitoring programs.

Long term monitoring – allergenicity effects

As noted above, food is a complex mixture. Food allergies are caused by a wide variety of foods and can be considered as adverse reactions to otherwise normal foods or food components that result in an abnormal response of the body’s immune system to specific constituents of food (FAO-WHO, 2001). Although sometimes confused with metabolic intolerance, true food allergies can involve several types of immunological responses. The most common type of food allergies is known as immediate hypersensitivity reactions and are mediated by the allergen specific immunoglobulin E (IgE ) antibodies. IgE mediated responses can occur to a wide variety of substances including pollens, moulds, insect venoms and foods. Immediate hypersensitivity reactions are typically associated with symptoms within minutes, to a few hours, after exposure to the triggering food. IgE mediated reactions affect approximately 10-25% of the population, although food allergies are only a small proportion of the overall occurrence. This type of allergic reaction more frequently affects infants and young children than adults, and some authors have speculated that the prevalence of occurrence in infants under three years of age may be as high as 5-8% (FAO-WHO, 2001).

Food allergies may also manifest as delayed hypersensitivity reactions in which the response occurs more than eight hours following ingestion of the food. Delayed hypersensitivity reactions are well described in infants. Delayed hypersensitivity reactions have been observed in response to milk and soybeans, although the most common delayed hypersensitivity response in the population is celiac disease, which is believed to affect about 1 in 300 to 3000 people.

Food allergies can be caused by a wide variety of foods. The UN Codex Alimentarius Commission Committee on Food Labeling has established a list of common allergenic foods associated with IgE responses that includes peanuts, soybeans, milk, eggs, fish, shellfish, wheat and tree nuts. These common allergenic foods are believed to account for over 90% of all moderate to severe allergic reactions to food. Some authors have also suggested that certain classes of drugs, such as the $ blockers, non-steroidal anti-inflammatory drugs and the so-called ACE inhibitors that are widely used by adults and the elderly may predispose to the development of food allergic reactions as well (Wal, 2001).

Overall, the need for post market monitoring of allergenic foods, and food components, is made all the more critical by the lack of suitable, reliable and well defined animal models that can be used to reliably predict allergenicity prior to market introduction of a new, biotechnology derived, food or food component. Concern regarding the introduction of a novel, allergenic food can be considered on two levels. The need to prevent reactions in allergic consumers who are already aware of their allergy, and preventing de novo sensitization in genetically susceptible individuals who have not yet been identified as being allergic to a particular food or food component.

Top

As noted above, while post market monitoring of potential adverse health effects resulting from exposure to genetically modified foods has overall merit, the absence of adequate animals models that can reliably predict allergenicity makes this a particularly important post market tool. While post market monitoring is not a surrogate or substitute for pre market evaluation of a broad range of important health endpoints, including allergencity, it is likely that significant allergenicity to genetically modified foods, and components of genetically modified foods, may only be observed in careful, post market monitoring of human populations. Well-organized and properly conducted post market monitoring programs for allergenicity are important so that proper protocols can be implemented for the gathering of relevant medical histories. Properly conducted post market monitoring programs will also serve to reduce the inaccurate and unreliable reporting of food allergies in non supervised programs, which some authors have suggested could overestimate the true incidence by up to a factor of two (Wal, 2001). The development of post market monitoring programs for the detection of potential allergenicity has merit, but several methodological impediments must be overcome. It is generally recognized that while food allergies can be severe, they generally occur at a relatively low frequency. Moreover, the absence of reliable information on the precise background frequency of food allergies, particularly to conventional foods, in the general population further complicates the likelihood of detecting allergies that are specifically referable to exposure to foods, or food components, from genetically modified plants. Monitoring programs must therefore carefully define the size of the population at risk, as well as the usual hallmarks of age, sex and geographic location. Observations of workers who may be occupationally exposed in food production plants that are processing foods of genetically modified plant origin may also provide very useful information about allergenic potential. A US EPA Panel investigating the recent StarLink® corn dilemma in the United States noted that perceived food allergies, particularly in the case of high profile, well publicized food scares, may be ten times greater that the actual confirmed allergic population. In order to correctly capture the frequency and severity of responses in the population, great care must be taken to develop a systematic approach to verify the diagnosis of food allergy, as well as a mechanism to clearly identify the food responsible for the response, the quantity ingested, a clear description of the symptoms and timing of the allergic response and any previous and relevant clinical history. Ideally, confirmation of the response through determination of IgE and skin tests and/or food challenges should be considered desirable elements of a post market monitoring protocol.

A key element of a monitoring program resides in the ability to trace the origins of the genetically modified food, and all products derived from the modified food, along the entire food production chain. While the ability to trace foods along the entire food production chain is indeed a critical element of a monitoring program, it is also an elusive component as well. The great number of products that can be used as foods, incorporated as major ingredients (labeled) in final food products, or present at very low levels as food additives in processed foods where information or precautionary labeling would not normally occur makes it extremely unlikely that exposure to components of genetically modified foods in the general population is possible. Similarly, avoidance of exposure to such products by motivated consumers appears to be equally improbable as well. In summary, it is widely recognized that the pre-market animal models may not adequately predict potential allergencity in the human population. In addition, due to wide genetic variability within the human population, and for different geographical dietary intake, monitoring of unexpected potential adverse health effects can only be effectively considered once the product has been introduced into the market place. Post market monitoring of potential adverse health effects has been considered by a joint FAO-WHO expert panel (FAO-WHO, 2001). The Panel opined that a monitoring program should include a self-reporting system for any adverse health effects observed in consumers and those occupationally exposed in the food production industry. The Panel went on to note that reported data should be validated with respect to the clinical outcome in relation to allergy and the causality between the reported adverse effect and exposure to the specific genetically modified food or food component. The data obtained should be recorded, consolidated and published so that national regulatory and monitoring authorities internationally could learn and benefit from each other’s experiences. The Panel noted that the success of a post market-monitoring program would be directly related to the traceability and labeling of the genetically modified food or food component, the availability of background data on prevalence and incidence of food related allergies, knowledge of the existence of confounding food and non-food factors, knowledge of the changes in dietary patterns over time and the availability of trained experts, and the necessary infrastructure, to monitor, detect, record and react to food allergies that are likely related to genetically modified foods of plant origin.

Top

The role of labeling in long-term health effects monitoring

In its review of the safety of genetically modified foods, the expert panel convened by the Royal Society of Canada noted that a major issue in the public debate has focused on the labeling of foods derived from genetically modified plants (RSC, 2001). The Panel unanimously supported the concept of mandatory labeling of genetically modified foods, where practical and possible, but only in those instances where there are clear and scientifically established health risks posed by the novel food product. The Panel opined that there are several instances where such clear evidence would exist, including those cases where there are well-established uncertainties in the identification or evaluation of certain health risks. Given the uncertainties associated with potential allergencity and other potentially significant long term health effects described elsewhere in this Paper, mandatory labeling in such instances would seem reasonable. Notwithstanding, even in such cases, the requirement to label would be rooted in a desire to inform the consumer of the uncertainty in the safety evaluation process and the need for sensitive consumers to be prudent. It is important to recognize that the Panel, in this instance, considered mandatory labeling to be an issue of "informed choice" or a means of engaging the general public in the risk management process, a position not unlike that frequently adopted internationally (Codex Alimentarius Commission, 2001), and not an instrument for long term monitoring of potential adverse health effects. Similarly, the Panel also noted the general lack of knowledge about the overall safety of conventional foods, and concluded that there would be little scientific justification for regulating the labeling of both types of foods differently.

In conclusion, and as noted by the Panel, there currently does not exist a sufficient scientific basis on which to recommend mandatory labeling of genetically modified foods, either for the purpose of informing consumers or as a vehicle that could be employed in the long term monitoring of potential adverse health effects.

Application of the "Precautionary Principle" to regulation and long term monitoring

Few topics in the public debate of potential harmful substances in the environment have evoked as much controversy or discussion as the application of the "precautionary principle" to this issue.

The concept of a "precautionary principle" can be traced to the 1992 Maastricht Treaty of the European Union and Principle 15 of the 1992 United Nations Rio Declaration on Environment and Development. Principle 15 states, inter alia, that where there are threats of serious or irreversible damage, lack of full scientific certainty should not be used as a reason for postponing cost-effective measures to prevent environmental degradation.

Some have interpreted the Principle to mean that we should ban all activities about which we know little on the chance that they may, one day, be shown to be harmful to the environment or humans. The simplified application of the Principle in the context of its likely intended use would suggest that we should not wait for absolute scientific certainty before acting (banning a substance or activity) if the activity or substance is believed to cause widespread, serious, and irreversible risk of harm.

The case in point, on which the Principle was built, was the CFC refrigerants and ozone depletion. These substances were being released into the environment in large amounts and were spreading globally. On reaching the stratosphere, they contributed to the destruction of the ozone layer with resultant increases in UV radiation at the surface. Because of their persistence, effects on the ozone layer would likely require millennia for recovery. The increased UV posed a potentially serious global threat to human health and the environment. The effects were thus serious, widespread and only very slowly reversible. The decision to restrict these substances was not made in the absence of data. Scientists had been studying the holes in the ozone layer for years and considerable knowledge was available.

Do genetically modified foods fit the "precautionary principle" paradigm? Firstly, these substances are well studied and the evidence of harm from the present array of genetically modified foods is not compelling. The "harm" that they may produce to the human health is likely minimal, especially when one considers only those genetically modified organisms and foods in commerce today. The nature of the scientific method to ascertain the safety of genetically modified foods, indeed of any food – genetically modified or otherwise – encompasses elements of uncertainty. While premarket safety evaluation of genetically modified foods cannot provide absolute certainty regarding their safety, and while available methodology for post market long term monitoring and surveillance will not contribute significantly to the resolution of the scientific uncertainty, application of Principle 15 of the Rio Convention — the Precautionary Principle — to genetically modified foods is not appropriate.

Top

Bibliography

Canadian Cancer Society, 2001. Canadian Cancer Statistics. A Joint Report of the National Cancer Institute of Canada, the Canadian Cancer Society and Health Canada. (Available on line: www.cancer.ca)

Codex Alimentarius Commission, 2001. Joint FAO/WHO Food Standards Programme. Report of the Twenty-Ninth Session of the Codex Committee on Food Labelling. Ottawa, Canada

Codex Alimentarius Commission, 2000. Joint FAO/WHO Food Standard Programme, Codex Ad Hoc Intergovernmental Task Force on Foods Derived from Biotechnology, Chiba, Japan, March 2000.

FAO-WHO, 2001. Evaluation of Allergenicity of Genetically Modified Foods. Report of a Joint FAO/WHO Expert Consultation on Allergenicity of Foods Derived from Biotechnology. Rome, Italy

FAO-WHO, 2000. Safety Aspects of genetically Modified Foods of Plant Origin. Report of a Joint FAO/WHO Expert Consultation on Foods Derived from Biotechnology. Geneva, Switzerland.

Health Canada, Adverse Drug Reactions Bulletin, 2000

Health Canada, News Release 2002-17, The Honourable Anne McLellan , Minister of Health, Introduces a Bill to Modernize Pesticide legislation and Protect Canadians. March 21, 2002

Mausner, J.S. and S.Kramer, 1985. Epidemiology, An Introduction. W.B. Saunders Company, Philadelphia.

RSC, 2001. Elements of Precaution: Recommendations for the Regulation of Food Biotechnology in Canada. An Expert Report on the Future of Food Biotechnology prepared by the Royal Society of Canada.

US SOT, 2002. Society of Toxicology Position Paper on the Safety of Foods produced Through Biotechnology.

Wal, J-M. 2001. Topic 9: Post market Surveillance of Allergenicity. Joint FAO/WHO Expert Consultation on Foods Derived from Biotechnology. Rome, Italy