VI.180 Dr. McLean commented that there was a wide range of combinations and doses available for a range of types of cattle. The combinations were designed to ensure maximum dose response in the target animals. These combinations had been subjected to studies to ensure that the residue levels did not exceed prescribed values.

VI.181 Dr. Ritter indicated he had no expertise in this area.

Question 19:

What differences exist between the therapeutic or zootechnical use of hormones permitted by the EC and the use of hormones for growth promotion purposes, as permitted in the US and in accordance with good animal husbandry practice and/or good veterinary practice, in terms of administered quantities, residue levels and potential adverse effects on human or animal health? Do common therapeutic or zootechnical uses of the hormones in question in the EC involve large scale (ie.,whole herd) or repetitive treatments?

VI.182 Dr. André observed that the use of the hormones for therapeutic or zootechnical purposes and for growth promotion purposes was not comparable for many reasons. The number of animals treated was different, as for therapeutic use very few animals were treated. For zootechnical purposes, the number was higher but remained limited. There were no systematic, repetitive treatments. For growth promotion purposes, the number of animals seemed to be large.

VI.183 In addition, Dr. André stressed that the future of the animals treated either for therapeutic or zootechnical purposes or for growth promotion purposes was very different. In the first case the animals usually remained on the farm for months or years after treatment and the problem of residues was not of concern. The only comparable situation was for oestradiol benzoate, for which, in case of unsuccessful treatment in a cow, a withdrawal period of two months had to be observed. In the second case, the animals were slaughtered weeks or months after the treatment. Furthermore, the delivery conditions differed. In the EC, the therapeutic use of these hormones was particularly regulated. Finally, Dr. André noted that none of these therapeutic/zootechnical uses would change the level of residues of hormones in meat on large scale as occurred from growth promotion use. No scientist had said that it was a bad thing to use these hormones for therapeutical use; their use on a large scale for growth promotion was not the same thing.

VI.184 Dr. Arnold replied that the total dose contained in an implant of oestradiol-17b was much higher than the dose injected for therapeutic purposes because the implant contained the quantity needed over a long time (e.g., 200 days). On a daily basis, however, the dose injected was much higher, and therefore, tissue residues (excluding the injection site) after treatment were much higher if compared with the residues measured at any time after implantation. No representative data were available for progesterone or testosterone that would permit comparisons. With the exception of oestrus synchronisation, individual identified animals were treated.

VI.185 Dr. Lucier indicated that it was possible that extraordinarily low residues (likely non-detectable) of the hormones could be present at slaughter if the animals had received the hormones for therapeutic use (see response to question 3).

VI.186 Dr. McLean replied that the therapeutic use of hormones generally required the administration of a larger dose, which in turn resulted in greater levels of residues. The persistence of these residues depended on the dose, formulation and site of administration. Individual animals or groups of animals were often treated at one time, but rarely the whole herd. Treatments may be repeated. Treated animals should be identified so that they were not slaughtered for human food until a suitable withdrawal period had elapsed. It was often necessary to withhold milk. When hormones were used for growth promotion, the implants slowly released the active ingredients and there was never a high peak of hormones in the tissues. In the case of therapeutic use, there often was a significant peak.

VI.187 Dr. Ritter responded that the primary difference between the therapeutic and/or zootechnical uses of the hormones permitted by the Community when compared to growth promoting uses related primarily to the frequency of use, dose, and the potential number of animals to be treated. In all cases, therapeutic and/or zootechnical uses would generally involve lower doses at less frequent intervals in fewer animals, and, in the case of the Community, was restricted to administration by a licensed veterinarian. Notwithstanding, at least in the case of the natural hormones, the approach adopted by the Community for therapeutic/zootechnical uses provided for residue levels which fell within the range of levels normally found in untreated animals. While the basis for use of these hormones for growth promotion purposes was fundamentally different, the resulting residues appeared, nevertheless, to also fall well within the normal physiological range.

Question 20:

Do you consider that the conditions imposed by the EC for the therapeutic or zootechnical use of these hormones can achieve its aim of avoiding any potential adverse effects on human health resulting from residues in meat from animals treated for such purposes?

VI.188 Dr. André commented that this question was not related to a scientific expertise, and as a consequence, he could only give a personal opinion. When used for therapeutic or zootechnical purposes, these hormones were used by (or under the control of) veterinarians. This guaranteed that they were used according to the indications for which they had been registered (i.e. dosage, route of administration, withdrawal period). Moreover these hormones were used either on individual animals or in well defined herds, usually in order to synchronize oestrus. In this case, the animals were reproducting animals, they had a high value and would be slaughtered only several years after the treatment. The problem of residues was not of concern for these animals. It could not be compared with the use of these hormones for growth promotion. When used for therapeutic purposes, these drugs were needed to restore health and the benefit/cost ratio was very high. When used for zootechnical purposes, the consequence for a great majority of the animals was to become pregnant and, exceptionally, to be slaughtered.

VI.189 Dr. Arnold indicated that the products which could be used were reviewed according to the criteria established for the approval process of veterinary medicinal products, including established withdrawal times. If the approved conditions of use were observed, any potential adverse effects on human health resulting from residues in animals treated for such purposes could be avoided.

VI.190 Dr. Lucier replied that the EC measures for the therapeutic or zootechnical use of these hormones could not guarantee "zero risk", although the potential risks were likely lower than risks from their use as growth promoter.

VI.191 Dr. McLean responded that he was not fully familiar with practices in the Community. However, studies on antibiotic residues have shown that the involvement of veterinarians in the control and prescribing of antibiotics on farms did not always guarantee that residues would be controlled. The same situation would probably exist with hormones.

VI.192 Dr. Ritter recalled that the Community allowed the administration of oestradiol, testosterone and progesterone to animals for therapeutic and zootechnical purposes only under conditions which restricted administration to a veterinarian, required registration of treatments and a sufficient withdrawal period. Given the nature of conditions imposed for such use, it was unlikely that residues resulting from exogenous application of the gonadol hormones would be present as residues. Notwithstanding, growth promotion uses of the natural hormones similarly would not result in residue levels in excess of those expected in untreated animals.

Question 21:

Is there a difference in terms of potential adverse effects on human or animal health between residues of these hormones administered as feed additives, compared to residues of these hormones which are implanted or injected?

VI.193 Dr. André indicated that as far as he knew, the metabolism of these hormones administered as feed additives had not been studied. To produce residues meant that they were absorbed, and that they are efficient and metabolised. Nevertheless, their use as feed additives could induce a risk for people potentially in contact with these hormones (factories, farmers, children etc.), and a risk for the environment.

VI.194 Dr. Arnold responded that melengestrol acetate was only used as a feed additive. Zeranol and trenbolone acetate were used as implants only. Oestradiol­17b, progesterone and testosterone were administered by different routes including injection and implantation. It was difficult to make comparisons about potential effects due to means of administration because there were too many variables in the equation (chemical nature, dose, route, formulation, target species). If all uses were carried out in compliance with the established conditions of use, the remaining residues could be regarded as safe. The safety margins might vary from one condition to the other and the consequences of non­compliance might also be different.

VI.195 Dr. Lucier commented that the residues would be the same in terms of chemical structure, although the residue levels could be different quantitatively.

VI.196 Dr. McLean observed that the control of the administration of hormones to animals as feed additives was by removing them from feed at a given time. In practice, because of carry-over in feed mixing plants and residues in storage silos and feeding stalls, there was not a sharply defined cut-off point. In the case of antibiotics, it was often necessary to use a dedicated mixing plant to avoid carry-over from treated to untreated feed. The dose absorbed could also vary depending on food intake and absorption from the gut. There was also the problem of accurately mixing small amounts of potent active ingredients through the feed. Injection was more accurate and reliable, and on balance the method of choice.

VI.197 Dr. Ritter replied that potential adverse effects on human health invariably related to residues which exceeded acceptable limits. Proper use of growth promoting hormones as either implants or feed additives might result in residues which, if used in accordance with recommended and approved practice, should not exceed acceptable levels regardless of the method or means of administration. He noted that improperly placed implants could result in failure to properly identify implantation sites at slaughter and could thereby result in excessively high residues in those tissues which would otherwise be identified and removed. Properly located implants, when implanted in accordance with approved practice, should not result in unacceptable residues and hence there should be little impact on potential adverse human health effects regardless of the method or route of administration.

Question 22:

With respect to Zeranol, Trenbolone and MGA, what are currently the technological limits of detection and/or quantitation of residues? Ie., what is the lowest level of residue which can be detected?

VI.198 Dr. André observed that these hormones were considered as classical banned compounds (within the EC). They were controlled with methods whose limits of detection (LD) were equal or better than 2 ppb. However, most methods in use for screening purposes of these hormones were immunological ones. Their real limits of detection were usually lower than 0.5 ppb. Mass spectrometric methods allowed confirmation of results at this level. Furthermore, using techniques with high performance output such as high resolution mass spectrometry or MS-MS, the level of 10 ppt. or so could be reached.

VI.199 Dr. Arnold replied that the technological limitations would allow the quantification of certain residues in certain tissues at concentrations below 1 ng/kg. This could not, however, be achieved in routine monitoring. Moreover, the limits of detection of current routinely applied methods may vary from country to country (within certain limits). Only few methods had been internationally collaboratively studied according to internationally harmonised protocols. Only few methods were fully validated in accordance with the criteria first defined by the Codex Alimentarius and later adopted by the EEC in Directive 85/592.

VI.200 Dr. McLean explained that there were two basic types of assays used. Screening methods were used to detect drugs for metabolites at the level of interest. These had high sample throughput and aimed to avoid false negative results. Confirmatory methods provided unequivocal identification of the drug and/or metabolites at the level of interest. The cost of analysis was considerably greater than for screening methods. These methods were constantly being reappraised.

VI.201 Dr. Ritter responded that methodology for MGA had been reported³⁰⁸ with a limit of detection of 5 ppb. and a limit of quantitation of 10 ppb., both in fat, the target tissue for this residue. LC methodology for the trenbolone acetate had been reported³⁰⁹ with a limit of detection of 2 ppb in muscle and liver for b-trenbolone, and a limit of quantitation of 2 ppb in muscle and 4 ppb in liver for a-trenbolone in liver and kidney. GC/MS methodology for zeranol had been reported by Covey et al (1988) with a limit of detection of 0.1 ppb. and a limit of quantitation of 0.2 ppb. in liver or muscle. The Joint FAO/WHO Expert Committee on Food Additives (WHO, 1988) reported that radioimmunoassays could detect free and conjugated a- and b-trenbolone at levels of 75ng/kg in tissues. Radioimmunoassays were generally regarded as appropriate screening methods only, while conventional analytical procedures such as gas chromatography/liquid chromatography/mass spectrometry are generally regarded as suitable for confirmatory analysis.

Question 23:

Can you describe how the EC ban is enforced in the EC's internal market both formally and in practice? How is the EC ban enforced at the EC's borders both formally and in practice? How are the conditions linked to the exceptions to the ban (i.e. administration of hormones for therapeutic and zootechnical purposes) enforced, both formally and in practice, in the EC's internal market and at the EC's borders? What is the formal content and practical effect of the EC's control programme which ensures that imported products do not receive treatment more favourable than domestic products?

VI.202 Dr. André noted that this question was not related to scientific expertise. From a personal point of view he believed that the important number of workshops, international congresses (e.g. Congress on Anabolizing Agents in Gent : 1992, 1993, 1994; Euroresidue I in 1990, Euroresidue II in 1993, Euroresidue III in 1996), laboratory networks system, European research programmes, as well as the development of Quality Assurance Systems in Europe had to be considered.

VI.203 Dr. Arnold replied that since its entry into force Council Directive 86/469/EEC had essentially harmonised the control of residues in live animals and meat. Starting from 1987, EC member States presented an annual updated national residue control plan for approval by the Commission. The results of residue controls were also reported to the Commission. The Directive was very clear and relatively inflexible in what concerned the identification of the substances and the species of animals to be checked and the number of samples to be taken. Unfortunately, a major part of the analytical capacities of the laboratories of the EC member States had to be used regularly to monitor substances which did not necessarily present the real problem areas (e.g., the stilbene oestrogens).

VI.204 The scientific concept behind sampling, however, was not clear and was subject to different interpretations. It seemingly mixed elements of a statistically based monitoring system with other concepts. The reporting format was only semi­quantitative with respect to the reported residue contents of positive samples. In consequence, the results could not be used to compare the situation in the EC member States or, for example, to conduct a crude assessment of exposure of European consumers to residues of anabolic substances.

VI.205 Dr. Arnold further explained that a network of reference laboratories had recently been set up with an outstanding European reference laboratory on top. However, the moderate financial contribution to the European reference laboratories from the Commission was given only on an annual basis, leading to difficulties in ensuring experienced permanent staff in these institutes. The quality of the national reference laboratories co­ordinating the activities of the routine laboratories carrying out the daily work could not be judged. Although it was not known how many routine laboratories were accredited nationally, it was obvious that systems of quality management had not yet been established everywhere in the Community.

VI.206 The endogenous production in the animals of oestradiol­17b, testosterone and progesterone causes difficulties in the control of their use. MRLs have not been set, and cannot be set, for the therapeutic and zootechnical uses, because the distribution of the three natural hormones in the three categories of animals (untreated animals, animals illegally treated for growth promotion, and animals legally treated for the permitted purposes) largely overlap. Because withdrawal times could not be enforced on a broad scale and distribution chains were insufficiently controlled, only limited attempts ­ requiring a lot of man­power and other resources ­ could be made to enforce compliance. If there was a suspicion, of course, the animal involved can be traced back and the producer questioned about, e.g., the prescription of the drug, the veterinarian involved etc. The main difficulty was to identify suspect animals in order to start further investigations and measures. Only the discovery of an injection site or an implant containing illegal substances could provide proof of a treatment. Decision limits for the levels of oestradiol­17b had provisionally been established in order to identify suspect animals.

VI.207 Enforcement was difficult if people did not want to comply with the legal rules ­ which otherwise would perfectly guarantee consumer protection. It appeared that illegal anabolic hormones were readily available. The expectation of extra profits was an incentive for the continued the use of these substances including beta agonists, not only for farmers but also for participants in the illegal network of distribution.

VI.208 Dr. Arnold recalled that the requirements of Directive 86/469/EEC also applied to third countries exporting live animals and meat to the Community. Missions were regularly sent to these countries to evaluate their systems of residue monitoring and surveillance. At the border, enforcement relied entirely on the discriminating power of analytical methods. This was not possible in the case of meat from animals treated for therapeutic or zootechnical purposes, because this meat was not different from meat obtained from animals never treated or treated with the same substances for growth promotion. Although domestically produced and imported products were formally treated equally, the possibility of detecting illegal practices of growth promotion (illegal referring to non­compliance with EU legislation, not to uses which were also prohibited in the country of origin) was probably significantly greater for domestically produced meat because of the possibilities to inspect production plants with a history of non­compliance, to take samples from live animals, etc.

VI.209 Dr. Ritter indicated that he had no expertise with regard to EC enforcement, in relation to a discussion about the number of animals checked per year in the European Communities and other countries. However, he noted that the number of samples taken could be misleading. The intent of any monitoring programme was the development of a programme that had the statistical confidence necessary to detect violations to the extent that the compound was used, to identify residue levels which were out of compliance. The relevant question was whether the monitoring programme was statistically confident and able to detect compliance. Most monitoring programmes were intended to detect a 5 per cent violation rate 95 per cent of the time. The number of samples that were required to do that varied from country to country and from commodity to commodity because it was a function of the use practices for the substances involved.

Question 24:

What analytical methods, or other technical means, of residue detection exist to control the use of the six hormones in dispute for growth promotion purposes in accordance with good animal husbandry practice and/or good veterinary practice. What means exist to control the use by farmers of the six hormones in dispute for growth promotion purposes in accordance with good animal husbandry practice and/or good veterinary practice? What are their respective cost implications?

VI.210 Dr. André replied that since these hormones were prohibited within Europe, their control was included in national plans, the aim of which was to survey their misuse. Analytical methods were available both for screening purposes and confirmatory purposes. For screening purposes, several methods had been developed. Research was also carried out to prove the exogenous origin of natural hormones, when detected in urine samples, based on methods used in human for doping control. For confirmatory purposes, different methods existed, with most of the available methods being based on GC-MS detection and identification. These techniques had been developed for matrices like urine, fat, tissue, faeces or injection sites.

VI.211 Dr. Arnold noted that the analytical community of the European Communities was very active. To provide the appropriate methodology to detect illegal practices of growth promotion ­ particularly the potentially hazardous practices involving black market drugs ­ was a great challenge for the scientists working in both academic and official laboratories. As a result, a large number of probably valid ­ although not always validated ­ methods relying on different principles was available. Furthermore, in the European Communities it was possible to inspect farms and sample live animals, with the consequence of fines or legal prosecution in case of violations.

VI.212 Dr. McLean indicated that descriptions of analytical methods use were published by the regulatory authorities of many countries. Control of use by farmers was done through legislation which detailed restrictions on the sale, possession and use of the various hormonal preparations. The cost of these measures varied from country to country and the cost of analysis depends on the drug in question and whether screening or confirmatory methods were employed.

VI.213 Dr. Ritter responded that the residue detection methodology utilized to control the use of the six growth promoting hormones generally followed those procedures described in question 22. Detection methodologies for veterinary drug residues were largely described by two main types of analysis (Blanchflower, 1995). Screening methods generally had a high sample throughout, permitting analysis of large numbers of samples in a relatively short period of time, and at minimal cost. They generally had a low probability of false negatives and could detect, but not accurately quantify, potentially positive samples. Screening methods for the growth promoters were typically represented by the radio immunoassays. In contrast, confirmatory methods were used to confirm unequivocally, and to quantitate, the presence of a drug residue. They were characterized by relatively low throughput and high cost both in terms of equipment and supplies, and a low probability of false positives.

VI.214 Control of veterinary drugs used in food production by farmers was generally achieved by national authorities through imposition of regulations requiring that drugs of this type be dispensed only on the written authority of a properly qualified veterinarian and in compliance with relevant use information established by appropriate national regulatory authorities. Compliance was further assured through rigorous monitoring and, where appropriate, through enforcement activity. Cost implications generally related to the increased cost associated with the requirement that such drugs only be administered under the authority of a properly qualified veterinarian. Costs would also be proportional to the rigour with which a monitoring and compliance programme was designed and implemented.

Question 25:

What analytical methods, or other technical means, exist to control the use of the six hormones in dispute according to the conditions set out in the proposal of the EC Commission in 1984 (COM(84)295 final) and what are their respective cost implications?

VI.215 Dr. André noted that this twelve year old proposal had been rejected by the European Parliament.

VI.216 Dr. Arnold recalled that this proposal had envisaged the controlled use of the three natural hormones for growth promotion and proposed re­visiting the prohibitions of trenbolone acetate and zeranol after a scientific evaluation of these substances. In analytical terms, the enforcement problems would have been similar to those of today because residues in meat remaining after the (proposed) legalised uses for growth promotion would not have differed significantly from residues found after illegal uses in most cases. The pharmaceutical industry could have applied for marketing authorization of strictly formulated products, which would have been evaluated and approved under the veterinary medicines directives. There would have been competition between high­quality, efficacious and safe products and products of the black­market. However, Dr. Arnold could not judge whether this would have limited the growth of the now existing black market, thereby reducing the costs of enforcement.

VI.217 Dr. McLean noted that the methods proposed in that Directive were similar to the methods now used in other countries. However, the passage of time had permitted better analytical technology and techniques and increased automation, often resulting in increased laboratory throughput at decreased cost per sample. Farmers were also much better educated about the need to avoid residues which exceeded the MRL. The Community would now be in a much better position to screen for residues in meat products than it was in 1984.

Question 26:

Are there other products (including veterinary drugs) commonly used in the production of meat and animal products which have comparable potential adverse effects on human or animal health as the six hormones in question? If so, what analytical methods, or other technical means, of residue detection exist to control their use in accordance with good animal husbandry practice and/or good veterinary practices. What technical means exist to control their use by farmers in accordance with good animal husbandry practice and/or good veterinary practice? How do the cost implications for controlling the use of these other products differ from those used to control the use of the six hormones in question?

VI.218 Dr. André responded that to the extent a compound was active as a drug, it could exert toxic effects. Drugs were submitted to regulation and were registered for precise indications, under precise conditions, including data on MRLs and definition of withdrawal periods (see also response to question 10). Several methods for control were available. Drugs could only be registered for use when a method designed for residue control existed. Networks of laboratories and control plans were set up for drugs according to the same scheme as the one used for illegal drugs.

VI.219 For products for which particular conditions of use were indicated, the control was usually in the hands of practitioners. New recording systems were under development (see response to question 1). Controls by official bodies (e.g. Veterinary Inspection Services) were effective in some countries. In other countries regulations were changing in order to apply EC Directive 96/23, allowing for more controls to be performed on the farm itself. This was particularly true for the hormones in dispute, as for other banned compounds (chloramphenicol, rhonidazol, etc.).

VI.220 Dr. Arnold indicated that a rather large number of active principles (probably a few hundred) were used as veterinary drugs in one or all countries involved in this dispute. Some substances were so active that their therapeutic doses were as low as micrograms per kilogramme of body weight of the animal. Other substances were used in both veterinary medicine and in crop protection, with some transfer of residues from animal feed via products of animal origin to the human consumer. Considerable progress had been made to regulate these substances on the basis of "tolerances" for residues or MRLs. Many of the underlying ADIs had been proposed by JECFA. More JECFA evaluations had indirectly been appreciated and found acceptance (e.g., in the MRL setting process within the European Union) as one would conclude from the list of substances which had formally passed the elaboration process of the Codex Alimentarius Commission.

VI.221 In the European Communities, no new active principle had been placed on the market since 1992 unless a Community MRL had been established under the provisions of regulation 2377/90. New animal drug applications had to be accompanied by a proposed regulatory method suitable to enforce tolerances/MRLs. The main performance characteristics (accuracy, precision, limit of detection, limit of quantification, specificity) and the ruggedness, practicability, applicability, susceptibility to interference of the methods were reviewed together with the other documentation relating to the quality, safety and efficacy of the drug.

VI.222 The review of the old substances, however, was still ongoing. The evaluation of the old substances had also begun in the pesticides sector. The majority of these substances had a long history of uses. Although they had never been evaluated according to contemporary safety requirements, they could probably be regarded as safe in the light of the experience with these substances. Should new evidence indicate that human health hazards could arise from the permitted uses of these substances, the competent authorities would withdraw the respective products until the problems were clarified. There were some examples, however, where highly efficacious substances with a long history of use had to be withdrawn from the market of veterinary drugs for food animals when more contemporary standards were applied in their re­evaluation (e.g., chloramphenicol, nitrofuranes, nitroimidazoles).

VI.223 When old substances were reviewed in the Community to establish MRLs, the Committee for Veterinary Medicinal Products always considered the available analytical methodology. An MRL was not proposed if no analytical method was available. The review of the methods and international harmonisation of methods suitable for the enforcement of Codex Standards was conducted by the FAO/WHO Codex Committee for Residues of Veterinary Drugs in Foods.

VI.224 The more difficult problem was to monitor so many substances in all foods of animal origin. The chemical properties of these substances were so different that it was impossible to include all residues in a few multi­residue methods. There were substances requiring special equipment for their detection. It was not practically feasible to monitor all potential residues in all food commodities every year. The most reasonable approach was to categorise compounds according to potential hazard and the likely exposure of consumers to the residues. Some compounds required inclusion in residue­monitoring programmes every year; others might be selected for longer cycles. For a few compounds, perhaps no monitoring was necessary. Directive 86/469/EEC covered veterinary drug residues in meat. Starting in 1988, EC member States and countries exporting meat to the Community had to submit residue control plans also for these substances. EC member States had more flexibility to adapt the plans to the actual needs (e.g., areas of risk). The new legislation adopted in 1996 enlarged the scope to cover other foods than meat. An increasing number of screening tests were now available which were based, e.g. on microbiological inhibition or immunochemistry. These methods were suitable to identify suspect "positives" (samples violating MRLs) and exhibit a low rate of "false­negative" results. However, an important aspect of the control of "good practices" was the control of distribution; distribution of veterinary drugs was not harmonised in the Community.

VI.225 Dr. McLean replied that many products used in animal production had the potential to cause adverse effects on human or animal health. The hazard was identified and managed through the use of "good practice", the application of an MRL, residue surveys and legal sanctions for violations (see also responses to questions 11, 22 and 24). There was no significant difference in the cost implications for controlling the use of other veterinary drugs when compared with the hormones.

VI.226 Dr. Ritter replied that there were a variety of products, including veterinary drugs and pesticides, which were utilized in the production of meat and animal products. Establishment of MRLs for other production aids, including both veterinary drugs and pesticides, implied that these chemicals could also be used essentially free of potential adverse effects in the human population, provided that appropriate conditions had been followed in the use of the production aid. Simply stated, once MRLs had been established, adherence to internationally accepted maximum limits implied that all products carry similar risks, or more appropriately stated, a similar lack of risk, as did the hormones at issue in this dispute.

VI.227 Control of use was generally accommodated through appropriate control of sale (restriction of use generally associated with "prescription" drugs) and rigorous monitoring and compliance programmes, further supported by appropriate enforcement action. In regard to analytical methodology available for detection and monitoring purposes, both screening and confirmatory methods were utilized. Availability of appropriate and reliable analytical methodology for other production aids, such as veterinary drugs, varied somewhat with the specific agent and might range from difficulties of bound residues with carbadox and nitrofuran, to the special and important considerations which were necessary when assessing the microbiological risk due to residues of antimicrobial drugs in food.

Question 27:

What potential adverse effects on human health arise from the occasional consumption of meat containing residues in excess of the Codex MRL in the case of Zeranol and Trenbolone?

VI.228 Dr. André noted that all of the previously described potential adverse effects on human health may arise. With regard to their hormonal effects, the lower the frequency and the level of residue, the smaller is the risk of health hazards. With regard to their potential carcinogenic effects, the frequency of consumption and the level of residue was not of concern.

VI.229 Dr. Arnold indicated that if the residues were due to treatment of the animals and not to contamination of the meat and the meat was not an injection site so that the causes of the violation of the MRLs were limited to biological variability of the animals response to treatment; he did not know of any specific adverse effect on human health which could arise from the occasional consumption of such meat because of non­compliance with the established conditions of use, and if this was a single, rare or occasional event. Any such event would, however, require regulatory action and ­ where indicated ­ legal prosecution in order to guarantee the integrity of the margin of safety which had been found to be required to ensure the medium and long­term protection of the consumers' health.

VI.230 Dr. Lucier responded that it would be necessary to conduct formal risk assessments to estimate exposure-effect relationships following different levels of exposure for the six hormones. In order to be useful, endpoints of concern needed to be determined (i.e., cancer, hormonal effects, etc.) and dose response models applied and/or developed that made use of all available and relevant information. It might be necessary to conduct additional research to put the risk assessments on a credible scientific foundation. It should be noted, however, that one would never have all the scientific information needed to remove all uncertainty in risk assessments. Priorities had to be agreed upon regarding which pieces of information are most critical. It was also important to note that selection of the risk assessment. model could influence the result. Threshold or linear models might give different results although if "mimicking of hormone effects" were determinative, then the issue of threshold was not relevant since physiological levels of the natural hormone exceeded any threshold that might exist (see response to question 8). The synthetic hormones could exert toxic effects not only related to their hormone action but also because of other structural/functional properties.

VI.231 Dr. McLean observed that the occasional consumption of meat containing levels in excess of the Codex MRL was of no health concern. The degree of excess would need to be defined, but the type of toxic response and NOEL would give some guidance. It must be remembered that the NOEL was derived taking into account studies involving administration for a lifetime, with the lowest effect level based on a response which could be as minor as a body weight change.

VI.232 Dr. Ritter replied that in view of several assumptions inherent in the models utilized to develop MRLs, which overestimated safety (NOEL, application of safety factors, assumption that MRL was always present and that dietary exposure to residues would occur daily for an entire lifespan), it is most unlikely that significant potential adverse effects on human health would arise from the occasional consumption of meat containing residues in excess of the Codex MRL for zeranol and trenbolone. It was, however, important to note that the potential for adverse effects increased as a function of both the frequency and exposure level regarding consumption of residues in meat in excess of the Codex MRL for both trenbolone and zeranol.

Question 28:

How would you assess the feasibility of labelling with respect to meat products from animals treated with growth-promoting hormones? How does this compare with the feasibility of labelling in other food safety contexts?

VI.233 Dr. André indicated that this question was not related to scientific expertise, but to political decision. However, such labelling would be consistent with the traceability concept which was rapidly being improved in European countries (ready to be implemented in France). However, this implied that an adequate control to qualify products would have to be established; in the case of residues of hormones, detection in meat did not stand in the way of feasible labelling.

VI.234 Dr. Arnold said that in his view it was not feasible to control labelling, since it could not be confirmed through laboratory testing whether an animal had ever been treated with growth promoters. If the objective was to verify that something had never been used, e.g., a pesticide, the situation was similar. The chances to detect food­irradiation, on the other hand, were much better. Also, if genetically modified organisms were involved in food production, it was sometimes possible to routinely detect the altered genetic material. There seemed to be no generally valid answer to this question.

VI.235 Dr. Lucier replied that meat products could be labelled to indicate that they were from animals treated with growth-promoting agents so that consumers would have as much information as possible to decide which meat to buy. If growth-promoting agents produced meat with lower fat content, a consumer with cardiovascular disease might specifically want to buy meat from a growth-promoted animal.

VI.236 Dr. McLean observed that the effective labelling of treated meat would necessitate identification of the animal at birth, the farm, sale, slaughter, processing and on to the consumer. The logistics of the process would be complex and make it difficult to regulate, especially in the case of processed meat products. The cost would far outweigh any perceived benefit. The same problem arose with the feasibility of labelling other foods, but because of the complex process for meat processing, it was much more expensive. In his view, the money spent on such a labelling process would be better spent on ensuring microbiological safety of food. Food contaminated with microbes and their toxins disabled or killed many people every year. The adverse effects of veterinary drugs or agricultural chemicals were rarely reported, but reports of food poisoning from microbial contamination were very common. For example, a recent outbreak of Escherichia coli food poisoning in Scotland made 400 people seriously ill and killed 17, while a similar outbreak in Japan infected 10,000 and killed 11 (Coghlan, A (1997) New Scientist 153 (2066)7).

VI.237 Dr. Ritter replied that the issue of labelling of food products was the subject of intense public debate, to which there was no easy resolution. The issue with regard to the feasibility of labelling in the case of the hormones could therefore relate as much to the objective of such a strategy as to its feasibility. Use of the natural gonadal hormones in meat production produced residue levels consistent with those which might typically be expected in meat products from untreated animals. Other than to fulfil a "right to know" objective, labelling in this case would not appear to provide information of value to consumers in a health related context. Particularly noteworthy was that as residue levels are comparable in both treated and untreated animals, and as exogenous hormones are, in any case, quantitatively and qualitatively indistinguishable from those endogenously present, it was unlikely that a labelling programme could be implemented at any practical level.

VI.238 In the case of the synthetic hormones utilized in growth promotion, at least two had been subjected to international review and safe residue levels had been recommended. The purpose of such a labelling programme would be unclear given that an international review had already concluded that residues at or below proposed levels did not constitute a risk to consumers, even under the exaggerated calculations and assumptions utilized in the development of MRLs.

Questions 29-34 were follow-up questions to those presented earlier, and the responses to the follow-up questions have been included with the responses to the earlier questions.

Question 35:

With reference to therapeutic and zootechnical use of the natural hormones at issue, do you believe that the EC Directive 88/299 „ensures that there are no residues left in meat for human consumption" (para 19, EC first submission in the Canadian complaint)?

VI.239 Dr. Arnold replied that no such insurance can be given. The meat would always contain residues even if withdrawal times were observed. The concept of "no residue" had been largely abandoned some 20 years ago when it became evident that "no residue" was a function of the limit of detection of the analytical method. All permitted substances, including the hormones, were regulated on the basis of an "acceptable residue" in the EEC. The "no residue" concept still applied to banned substances, but this was trivial. In the case of hormones it was acceptable if the sum of endogenous plus exogenous hormones was at or within physiological limits.

VI.240 The response of Dr. Ritter to this question is contained in his response to question 20 above.

TO CONTINUE WITH EC MEASURES CONCERNING MEAT AND MEAT PRODUCTS (HORMONES) COMPLAINT BY CANADA

³⁰⁸ Anderson and Fesser (1996).

³⁰⁹ Shih-Hsien Hsu et al . (1988).