J.L. Herrman is WHO Joint Secretary of JECFA, International Programme on Chemical Safety, World Health Organization, Geneva, Switzerland.
R. Walker is Emeritus Professor of Food Science, School of Biological Sciences, University of Surrey, Guildford, Surrey, United Kingdom.
The Joint FAO/World Health Organization (WHO) Expert Committee on Food Additives (JECFA) has been meeting regularly since 1956 to evaluate food additives, contaminants, naturally occurring toxicants and residues of veterinary drugs in food. To date, the Committee has evaluated more than 1 200 chemicals, including more than 20 contaminants.
JECFA provides scientific advice to the Codex Alimentarius Commission (CAC), primarily through two of its general subject committees, the Codex Committee on Food Additives and Contaminants (CCFAC) and the Codex Committee on Residues of Veterinary Drugs in Foods (CCRVDF). JECFA also provides FAO and WHO member countries with scientific advice on chemicals in food. Because its primary client is CAC, its priorities are generally established by CCFAC and CCRVDF.
To date, JECFA has evaluated three mycotoxins: aflatoxins B, G, and M; patulin and fumonisins. However, governments and CAC are becoming increasingly concerned about the presence of mycotoxins in foodstuffs, as evidenced by the presence of several additional mycotoxins on the CCFAC priority list: fumonisins, trichothecenes and zearalenone for evaluation, and ochratoxin A for re-evaluation. Zearalenone will be evaluated at JECFA's fifty-third meeting in June 1999, but the others are awaiting further data before they are placed on the agenda for evaluation.1 As with nearly all contaminants, the lack of relevant toxicological and epidemiological data is the primary impediment to the evaluation of most mycotoxins.
The contaminants ochratoxin A and patulin have both been evaluated twice by JECFA. They were re-evaluated at the forty-fourth meeting in 1995, when the previous provisional tolerable weekly intake (PTWI) of 0.1 mg per kilogram of body weight for ochratoxin A was confirmed and a provisional maximum tolerable daily intake (PMTDI) of 0.4 mg per kilogram of body weight was established for patulin. A traditional approach was used in these assessments in that no-observed-effect levels (NOELs) and lowest-observed-effect levels (LOELs) were identified and safety factors were applied to both.
Expression of the tolerable intake on a weekly basis is considered to be appropriate for contaminants that may accumulate within the body over a period of time, as is the case with ochratoxin A. On any particular day, consumption of food containing above-average levels of the contaminants may exceed that day's proportionate share of the weekly tolerable intake. The weekly assessment takes into account such daily variations, with the primary concern being prolonged intake of the contaminant. By contrast, a PMTDI was established for patulin because it does not accumulate in the body and only occasionally is apple juice (the primary route of intake) heavily contaminated.
Levels of ochratoxin A and patulin contamination of foodstuffs and potential intakes of these contaminants were considered by the Committee during their evaluations. However, this information was not included as an integral component of the risk assessment because it was assumed that there is no appreciable risk when intake is below the PTWI or PMTDI.
Aflatoxins were first evaluated at the thirty-first JECFA meeting in 1987. The Committee recognized that aflatoxin B1 is a well-known potent hepatocarcinogen in all the mammalian species studied, with a wide range of sensitivities. While good data were available on the relationships among aflatoxin levels, duration of exposure and carcinogenicity in animals, the available data on the association between aflatoxin exposure and primary liver cancer in humans were difficult to evaluate because of the large number of uncertainties in these studies, including inadequate data on the dietary intake of aflatoxins, the contribution of hepatitis B virus to the etiology of cancer, and cultural and dietary status and habits. The Committee concluded that the available scientific information was insufficient for determination of the extent to which exposure to aflatoxins contributed to the increased incidence of primary liver cancer in the populations that were studied.
In view of these uncertainties, the Committee was unable to establish a figure for a tolerable intake level. It urged that the intake of dietary aflatoxin be reduced to the lowest practicable levels so that the potential risk could be minimized, and it recommended that efforts should be made to limit the presence of aflatoxins in food to irreducible levels. An earlier Committee meeting had defined an irreducible level as "that concentration of a substance which cannot be eliminated from a food without involving the discarding of that food altogether, severely compromising the ultimate availability of major food supplies".
Subsequent to this evaluation, CCFAC began working on standards for aflatoxin levels in food and feed products, but it has been exceedingly difficult to reach consensus on the maximum levels that should be included in these standards. A major impediment to consensus is that the levels of contamination of foodstuffs vary widely around the world and, with the "irreducible level" recommendation, decisions are made solely on trade grounds.
With respect to trade, the perspectives of delegations differ profoundly. Delegations from countries in which aflatoxin contamination is not prevalent want standards that are based on low maximum levels because they do not wish to see the quality of their food supply degraded. Delegations from countries in which the climatic conditions make aflatoxin contamination a problem naturally wish to have standards in which higher levels of contamination are permitted so that they can trade their products on world markets. In addition, when stringent international standards are used, the populations of these countries are placed at higher risk because products with low levels of contamination are exported, leaving the more contaminated, lower-quality products for domestic consumption.
After a number of years of little progress, CCFAC asked JECFA, at its Twenty-sixth Session in 1994, to provide estimates of the toxicological potency of aflatoxins and to derive estimates of the potential risks for different human populations. In response to this request, aflatoxins B, G, and M were considered at the forty-sixth meeting of the Committee in 1996. However, the evaluation could not be completed at that meeting and it was carried over to the forty-ninth meeting in 1997, when a detailed evaluation was completed (WHO, 1998).
The Committee reviewed a wide range of studies, in both animals and humans, that provided qualitative and quantitative information on the hepatocarcinogenicity of aflatoxins. It evaluated the potencies of these contaminants, linked potencies to intake estimates and considered the impact, and overall risks, of hypothetical standards on sample populations.
It was recognized that aflatoxins are among the most potent mutagenic and carcinogenic substances known. Extensive experimental evidence from test species shows that aflatoxins are capable of inducing liver cancer in most of the animal species studied. Most epidemiological studies also show a correlation between exposure to aflatoxin B1 and an increased incidence of liver cancer, although there is some evidence suggesting that humans are at substantially lower risk from exposure to aflatoxins than test species are. Some epidemiological studies suggest that the intake of aflatoxins poses no detectable independent risk and others suggest that it poses risks only in the presence of other risk factors such as hepatitis B infection.
A number of factors influence the risk of primary liver cancer, most notably carriage of hepatitis B virus which is determined by the presence of hepatitis B surface antigen in serum. The potency of aflatoxin B1 appears to be significantly enhanced in individuals with simultaneous hepatitis B infection. This interaction makes it difficult to interpret the epidemiological studies and determine the extent to which aflatoxins act as independent risk factors.
The identification of hepatitis C virus is an important recent advance in understanding the etiology of liver cancer. Two studies have investigated interactions among hepatitis C infection, aflatoxins and liver cancer but, so far, the results have been inconclusive. It is estimated that between 50 and 100 percent of cases of liver cancer are associated with persistent infection with hepatitis B and/or hepatitis C.
The Committee considered that the weight of scientific evidence, which includes epidemiological data, studies in laboratory animals and in vivo and in vitro studies of metabolism, supports the conclusions that aflatoxins should be treated as carcinogenic food contaminants and that their intake should be reduced to levels as low as reasonably achievable.
A number of dose-response analyses have been performed on aflatoxins. However, all of these analyses have limitations, the most important of which are the following:
Observations relating to the interaction between hepatitis B infection and aflatoxins suggest two separate aflatoxin potencies; one is apparent in populations in which chronic hepatitis infections are common, the other in populations in which chronic hepatitis infections are rare. In analyses based on toxicological and epidemiological data, potency estimates for aflatoxins were divided into two basic groups, potencies applicable to individuals without hepatitis B infection and those applicable to individuals with chronic hepatitis B infection. As can be seen from the Figure, extrapolations of animal data to estimate potencies in humans generally fall within the range of potency estimates derived from the epidemiological data.
Note:HBsAg+/- refers to
presence or absence
of the hepatitis B surface antigen in serum.
Source: WHO, 1998
In deriving the potency estimates of 0.01 and 0.3 cases per year per 100 000 people, per ng of aflatoxin B1 per kilogram of body weight per day for hepatitis B surface antigen negative and positive individuals, respectively, the Committee used only those potency estimates from the epidemiological studies that showed a positive association between aflatoxins and liver cancer. Studies in which no association was detected, or in which the association was negative, were not used, leading to an overestimate of aflatoxin potency. Other potential biases included:
When these biases are taken into account, the values shown in the figure should be viewed as overestimates of the potency of aflatoxins, and it is possible that humans are less sensitive to aflatoxins than the animal species tested in laboratory experiments. Differences in the carcinogenic potency of aflatoxins among species can be partially attributed to differences in metabolism. However, there was insufficient quantitative information available about the competing aspects of metabolic activation and detoxification of aflatoxin B1 in various species to identify an adequate animal model for humans and to explain the apparent differences in potency among species.
In a population, the fraction of the total incidences of liver cancer that is attributable to intake of aflatoxins was derived by combining estimates of aflatoxin potency (risk per unit dose) and estimates of aflatoxin intake (dose per person). The frequency and amount of aflatoxin contamination in a number of products were available, and the Committee concentrated on groundnuts, cereals and maize. Many of the data on levels of aflatoxin contamination were derived from non-random samples that appeared to be biased upwards because monitoring studies focus on products that are thought to be contaminated. Some of the data on levels of contamination are not likely to be based on current CAC sampling recommendations for aflatoxins. Accordingly, data on levels of contamination should be interpreted with caution and used only to infer patterns of importance in setting standards and not to provide exact contamination estimates.
Mean dietary intakes of aflatoxins for various regions were estimated using regional diets from the Global Environment Monitoring System - Food Contamination Monitoring and Assessment Programme combined with data on levels of aflatoxin contamination.
The Committee considered the possible impact of applying two hypothetical standards to aflatoxin contamination - 10 and 20 mg per kilogram. These should not be construed as recommended standards; they were chosen for illustrative purposes only. Governments and CAC should perform this type of analysis for particular standards that are being considered, basing it on their own data on contamination, dietary patterns and prevalence of hepatitis B.
Two examples using these hypothetical standards were developed. In the first example, it was assumed that the level of contamination of food by aflatoxins is low and the proportion of the population that is hepatitis B surface antigen positive is small. Monitoring data from Europe on aflatoxin B1 levels in groundnuts, maize and their products were used and it was assumed that 1 percent of the population is hepatitis B surface antigen positive. Assuming that these foods are ingested according to the "European diet", the mean estimated intakes of aflatoxins are 19 ng per person per day when a standard of 20 mg per kilogram is applied and 18 ng per person per day with a standard of 10 mg per kilogram. Using the potency estimates shown in the Figure, these intakes translate into estimated population risks of 0.0041 and 0.0039 cancers per year per 100 000 people when standards of 20 and 10 mg per kilogram, respectively, are applied. Thus, reducing the hypothetical standard from 20 mg to 10 mg per kilogram yields a reduction in the estimated population risk of approximately two cancers per year per 1 000 million people.
The second example pertains to areas with higher levels of aflatoxin contamination and a larger percentage of the population carrying the hepatitis B virus. Monitoring data from China on aflatoxin B1 levels in groundnuts, maize and their products were used, and it was assumed that 25 percent of the population is hepatitis B surface antigen positive. Assuming that these foods are ingested according to the "Far Eastern diet", the mean estimated intakes of aflatoxins are 125 ng per person per day when a standard of 20 mg per kilogram is applied and 103 ng per person per day with a standard of 10 mg per kilogram. Using the potency estimates in the Figure, these intakes translate into estimated population risks of 0.17 and 0.14 cancers per year per 100 000 people when standards of 20 mg and 10 mg per kilogram, respectively, are applied. Thus, reducing the hypothetical standard for this population from 20 mg to 10 mg per kilogram yields a reduction in the estimated population risk of approximately 300 cancers per year per 1 000 million people. These calculations, and the basis for them, are explained in more detail in WHO, 1998 and WHO, 1999.
The differences in population risks between these hypothetical standards are lower than might be expected on a first analysis. However, the results are not surprising when it is considered that, in both cases, the most highly contaminated samples are eliminated, thus greatly reducing average estimated intakes of aflatoxins. The use of standards by all countries should, therefore, be encouraged.
After reviewing and analysing the data, the Committee came to the following conclusions:
FAO and WHO encourage governments and CAC to make use of this evaluation in deciding on the appropriate standards to apply to aflatoxins. Significant resources are required, however, as use of the evaluation requires a significant amount of information at the national level, including monitoring data and information on dietary patterns and the prevalence of hepatitis B in the population. A comforting factor is that risks are significantly reduced when the most highly contaminated samples are eliminated. This can be done without a detailed analysis of the risk, and governments are encouraged at least to take this step if they are lacking the information necessary to make full use of the evaluation carried out by JECFA.
1JECFA reports are posted on the FAO Web site (www.fao.org/waicent/faoinfo/economic/esn/jecfa.htm).
WHO. 1998. Safety evaluation of certain food additives and contaminants. Prepared by the forty-ninth meeting of the Joint FAO/WHO Expert Committee on Food Additives (JECFA). WHO Food Additives Series No. 40. Geneva.
WHO. 1999. Evaluation of certain food additives and contaminants. Forty-ninth report of the Joint FAO/WHO Expert Committee on Food Additives. WHO Technical Report Series No. 884. Geneva.