II.4.1.

A detailed and critical summary of the safety file shall be required.

II.4.2.

The detailed and critical summary shall:

II.4.3.

Annexes to the detailed and critical summary shall include:

II.5.1.

The data shall demonstrate that the substance has been precisely identified and characterised in order to ensure that the substance used in safety studies is reflective of the substance to be used in the field.

II.5.2.

Batches used in safety studies shall be identified and adequate specifications shall be provided, including purity (concentrations of impurities), isomer ratios and enantiomers, solubility and any other factor that may influence its activity.

II.5.3.

Information on the chemical and physicochemical properties of the substance may allow concerns to be identified and/or addressed based on known properties of substances with similar chemical and physicochemical properties.

II.6.1.1.

Data from the pharmacodynamic studies shall aim to enable the identification and characterisation of the mode/mechanisms of action that underlie the intended therapeutic effects as well as those underlying adverse effects/side effects. These studies shall be designed on a case-by-case basis taking account of available information with regard to the likely pharmacological actions for the substance.

II.6.1.2.

Particular consideration shall be given in relation to pharmacodynamic effects of the substance that may occur at doses below those required to produce toxicological effects, with consideration given to the need for derivation of a pharmacological ADI.

II.6.1.3.

Studies relevant to the establishment of a pharmacological ADI shall identify or characterise the mode of action, the dose-response relationship and identify a NOEL or BMDL, where possible and shall be used as a starting point from which a pharmacological ADI is derived. Where appropriate data are available from studies in humans (e.g. for substances with a history of use in human medicine) these shall usually be the most useful in identifying a pharmacological NOEL or BMDL. Guidance published by the Agency on the establishment of pharmacological ADI ( 6 ) shall be followed.

II.6.1.4.

Data on the pharmacodynamic effects of a substance shall:

II.6.1.5.

If pharmacodynamic data are not provided, their absence shall be scientifically justified and the impact of their absence discussed.

II.6.1.6.

If a pharmacological ADI is not derived, its absence shall be scientifically justified.

II.6.2.1.

Pharmacokinetic investigations shall provide information on the absorption of the substance, its distribution and persistence in the tissues, its metabolism and excretion. The oral route shall be the main route of administration in the pharmacokinetic studies as this is the route by which consumers are exposed.

II.6.2.2.

Metabolites produced in the laboratory animal species shall be compared to those seen in the target animal species, in line with the guidance provided in the International Cooperation on Harmonisation of Technical Requirements for Registration of Veterinary Medicinal Products (‘VICH’) — VICH GL47: Studies to evaluate the metabolism and residue kinetics of veterinary drugs in food-producing animals: laboratory animal comparative metabolism studies  ( 7 ).

II.6.2.3.

The pharmacokinetic data obtained from studies in laboratory animals shall be used to model the fate of the substance ingested by humans.

II.6.2.4.

Pharmacokinetic data in laboratory animals shall also be used to determine whether the metabolites that consumers will ingest in animal-derived food commodities are also produced in the laboratory animals used for safety testing. This is necessary in order to determine the relevance of the toxicological effects and NO(A)ELs or BMDLs obtained in the toxicology studies. If the laboratory animals produce the same metabolites to those produced by the food-producing animal, the laboratory animals shall be considered to have been auto-exposed to the metabolites that humans would consume. This is ordinarily taken as evidence that the safety of metabolites has been adequately assessed in the toxicology studies. If the metabolites produced by the target animal species are not produced in the laboratory animal studies, there may be a need to conduct safety studies using the major metabolite(s) produced in the target animal.

II.6.2.5.

Pharmacokinetic data may also help to explain unusual results obtained in toxicity studies, such as an apparent lack of dose-response when the drug is not well absorbed.

II.6.3.1.1.

Animal studies shall be performed by the oral route since this is the route of exposure for the consumer.

II.6.3.1.2.

Animal studies shall be conducted in established strains of laboratory animals for which historical data are available. Each substance shall be tested in the species and strain of animals that is the best model for its effects in humans.

II.6.3.1.3.

The substance to be tested shall be the active substance. However, if residues in foods derived from treated animals include significant amounts of a metabolite which is not produced in the laboratory animal species, the toxicity of the metabolite may need to be assessed separately.

II.6.3.1.4.

VICH GL33: Studies to evaluate the safety of residues of veterinary drugs in human food: general approach to testing  ( 8 ) shall be followed.

II.6.3.2.1.

Acute toxicity studies may have been performed for reasons other than the evaluation of consumer safety (e.g. for the evaluation of the user safety of a product) or may have been reported in published literature. Reports of any such studies shall be submitted as part of the Safety File.

II.6.3.2.2.

If available, acute toxicity data which may contribute to the overall picture of the toxicological profile of the substance and may highlight effects to look out for in longer term studies shall be provided.

II.6.3.3.1.1.

Data from repeat dose (90 day) oral toxicity studies shall be provided for both a rodent and a non-rodent species, together with the reasons for the choice of species, considering any available knowledge of the metabolism of the substance in animals and humans.

II.6.3.3.1.2.

Data from repeat dose oral toxicity testing studies shall:

II.6.3.3.1.3.

Guidance on the design of repeat dose (90-day) studies is provided in VICH GL31: Studies to evaluate the safety of residues of veterinary drugs in human food: repeat-dose (90 days) toxicity testing  ( 9 ) and shall be followed. Any departures from established guidance shall be justified and the impact discussed.

II.6.3.3.1.4.

The absence of repeat dose (90-day) oral toxicity studies in rodents and/or non-rodents shall also be scientifically justified and the impact of their absence discussed.

II.6.3.3.2.1.

Chronic toxicity testing shall be conducted in at least one species. This shall be the most appropriate species chosen on the basis of all available scientific data, including the results of the 90-day studies, with the default species being the rat.

II.6.3.3.2.2.

The data from chronic oral toxicity testing studies shall allow:

II.6.3.3.2.3.

Guidance on the design of repeat dose (chronic) studies is provided in VICH GL37: Studies to evaluate the safety of residues of veterinary drugs in human food: repeat-dose (chronic) toxicity testing  ( 10 ) and shall be followed. Any departures from established guidance shall be justified and the impact discussed.

II.6.3.3.2.4.

If a repeat dose (chronic) oral toxicity study is not provided, its absence shall be scientifically justified and the impact of its absence discussed.

II.6.3.4.1.

Data on tolerance in target species shall not be required for the evaluation of consumer safety. However, where relevant data have been generated or are reported in published literature, these shall be submitted as part of the Safety File.

II.6.3.4.2.

If available, data on tolerance in target species may contribute to the overall picture of the toxicological profile of the substance and may highlight effects to look out for in toxicity studies.

II.6.3.5.1.1.

General reproductive toxicity testing shall be conducted in at least one species, the default species being the rat. The oral route of administration shall be used.

II.6.3.5.1.2.

Tests for effects on reproduction shall aim to identify and characterise adverse effects of the test substance on reproductive performance of exposed adults as well as on the normal development of their progeny.

II.6.3.5.1.3.

Tests shall identify potential effects on male and female reproductive performance, such as gonadal function, oestrus cycle, mating behaviour, conception, parturition, lactation, weaning and on the growth and development of the offspring. These studies may also provide information about adverse developmental effects such as teratogenesis.

II.6.3.5.1.4.

If evidence suggests the occurrence of effects on development of the central nervous system, specific investigations of such effects may be required, for example through evaluation of results of other tests (see Section II.6.4.1).

II.6.3.5.1.5.

The data shall allow the establishment of a NO(A)EL or LO(A)EL or BMDL.

II.6.3.5.1.6.

Guidance on the design of reproduction toxicity testing studies is provided in VICH GL22: Studies to evaluate the safety of residues of veterinary drugs in human food: reproduction testing  ( 11 ) and shall be followed. Any departures from established guidance shall be justified and the impact discussed.

II.6.3.5.1.7.

If a reproduction toxicity study is not provided, its absence shall be scientifically justified and the impact of its absence discussed.

II.6.3.5.2.1.

The aim of developmental toxicity studies shall be to detect any adverse effects on the pregnant female and the development of the embryo and foetus as a result of exposure from implantation through the entire gestation period. Such effects may include enhanced toxicity in the pregnant females, embryo foetal death, altered foetal growth and structural abnormalities and anomalies in the foetus.

II.6.3.5.2.2.

If clear evidence of teratogenicity is seen in the rat, a study in a second species shall not be necessary except where a review of all the core studies indicates that the ADI would be based on the rat teratogenicity study. Testing in a second species (normally rabbit) is expected if no evidence of teratogenicity or equivocal results were seen in the rat.

II.6.3.5.2.3.

Guidance on the approach towards developmental toxicity testing is described in VICH GL32: Studies to evaluate the safety of residues of veterinary drugs in human food: developmental toxicity testing  ( 12 ). This provides for a tiered approach, with testing to be undertaken initially in a single species (rat). Any departures from established guidance shall be justified and the impact discussed.

II.6.3.5.2.4.

Studies shall use the oral route of administration.

II.6.3.5.2.5.

The data shall allow the establishment of a NO(A)EL or LO(A)EL or BMDL.

II.6.3.5.2.6.

If a developmental toxicity study is not provided, its absence shall be scientifically justified and the impact of its absence discussed.

II.6.3.6.1.

In most cases the substance to be tested shall be the parent compound only. However, in some cases there may be a need to test in addition one or more of the major metabolites separately. This would be the case if a major metabolite produced in the target species is not produced in the laboratory animal species.

II.6.3.6.2.

VICH GL23: Studies to evaluate the safety of residues of veterinary drugs in human food: genotoxicity testing  ( 13 ) identifies a standard battery of tests recommended for addressing the genotoxic potential of a substance. The standard battery includes tests aimed at detecting mutagenic, clastogenic and aneugenic effects. Any departures from established guidance shall be justified and the impact discussed.

II.6.3.6.3.

Results of genotoxicity tests shall be used to evaluate whether a substance is likely to cause genetic damage that may be passed from a parent cell to its daughter cells, either by direct or indirect effects on deoxyribonucleic acid (‘DNA’).

II.6.3.6.4.

Exposure to certain genotoxic substances is known to be associated with carcinogenesis and consequently, clearly positive findings in genotoxicity tests shall be considered to indicate that the substance may be carcinogenic. In addition, because germ cell mutations are known to be associated with disease, clearly positive findings in genotoxicity tests shall be considered to indicate that the substance may induce heritable disease (reproductive toxicity).

II.6.3.6.5.

The deliberate use of genotoxic substances that interact directly with DNA shall not be accepted in medicines for food-producing animals.

II.6.3.6.6.

The results from the genotoxicity tests shall contribute to the evaluation of the need for carcinogenicity data. Other factors that shall be considered in determining the need for carcinogenicity data shall be the existence of relevant structural alerts and the occurrence of pre-neoplastic findings in repeat dose toxicity tests.

II.6.3.6.7.

A substance that directly induces clearly positive findings in genotoxicity tests may only be accepted for use in food-producing animals if the genotoxicity findings are demonstrated not to be of relevance for the consumer. Results from carcinogenicity studies demonstrating the absence of neoplasia may form part of such a demonstration. Mechanistic data shall also be needed in order to demonstrate that the mechanism underlying the observed genotoxicity is not relevant for the consumer.

II.6.3.6.8.

In the absence of data to demonstrate that observed genotoxicity is not relevant for the consumer, clearly positive findings shall lead to the conclusion that an ADI cannot be established and that the substance is not appropriate for use in food-producing species.

II.6.3.6.9.

Clearly negative results from a standard battery of genotoxicity tests shall lead to the conclusion that the substance is not genotoxic.

II.6.3.6.10.

If equivocal results are seen in genotoxicity tests, the need for further testing shall be considered in light of the overall weight of evidence of the available data.

II.6.3.6.11.

In general the genotoxicity of major metabolites shall be considered to be adequately addressed by studies performed with the parent substance. However, if a major metabolite is produced in the target species but not in the laboratory animal species it may not be possible to conclude on the genotoxicity of residues without additional data generated using the relevant metabolite.

II.6.3.6.12.

In principle, identification of minor metabolites shall not be required.

II.6.3.6.13.

Minor metabolites are those present at levels below 100 μg/kg or that make up less than 10 % of the total residues as described in VICH GL46: Studies to evaluate the metabolism and residue kinetics of veterinary drugs in food-producing animals: metabolism study to determine the quantity and identify the nature of residues  ( 14 ).

II.6.3.6.14.

If the structure of a minor metabolite is known or can be hypothesised and if the metabolite is or is expected to interact directly with DNA, then the potential risk for consumers shall be addressed. Evidence shall be provided to confirm that its level is low enough to be considered virtually safe — i.e. the level shall be low enough to ensure that the increased cancer risk that would result from consumer exposure to the substance would be less than 1 in 106. This shall be achieved either using chemical specific data or, in the absence of such data, using the threshold of toxicological concern (‘TTC’) concept which provides an approach for quantifying the risk associated with a given exposure to a substance. Guidance published by the EFSA and the WHO on the TTC approach shall be followed ( 15 ).

II.6.3.6.15.

Similarly, if there is a concern that a minor metabolite present in food of animal origin is further metabolised in the consumer to produce a DNA reactive substance, evidence shall be provided to demonstrate that consumer exposure occurs at levels low enough to be considered virtually safe.

II.6.3.6.16.

For any of these substances (potentially genotoxic minor metabolites produced in the target animal or in the human consumer), the level of residues present in food of animal origin shall result in consumer exposure below the TTC at all time points following the start of treatment. As the possibility of exposure before the withdrawal period cannot be ruled out, and in light of the serious non-threshold based effect, it shall not be enough to demonstrate depletion to levels compliant with the TTC by the time point at which residues fall below the proposed MRLs.

II.6.3.6.17.

If more than one minor metabolite is DNA reactive, in the absence of evidence to the contrary it shall be assumed that all DNA reactive substances act by the same mode of action. Consequently, the total level of DNA reactive substances (dose addition) shall be compared with the TTC.

II.6.3.6.18.

Substances and metabolites that may cause cancer by mechanisms other than direct interaction with DNA may be assumed to have threshold based mechanisms of action. If such substances are to be used in veterinary medicines for food-producing animals, NO(A)ELs or BMDLs shall be established for the relevant effects in appropriately justified studies.

II.6.3.7.1.1.

VICH GL28: Studies to evaluate the safety of residues of veterinary drugs in human food: carcinogenicity testing  ( 16 ) provides guidance on factors to consider when determining the need for carcinogenicity testing and on carcinogenicity testing to be undertaken and this shall be followed. Any departures from established guidance shall be justified and the impact discussed.

II.6.3.7.1.2.

In those cases where carcinogenicity testing is deemed appropriate, the standard requirement shall be for a two-year rat study and an 18-month mouse study although, with appropriate justification, data from a single rodent species may be accepted.

II.6.3.7.1.3.

Genotoxic carcinogens shall not be accepted for use in food-producing animals.

II.6.3.7.1.4.

A substance that induces positive findings in carcinogenicity tests may only be accepted for use in food producing animals if the carcinogenicity data are demonstrated to be of no relevance for the consumer (for example, if the type of tumour seen is known to be of no relevance for humans) or if the carcinogenicity are demonstrated to be the result of a threshold dependent mechanism of action. In the latter case a NO(A)EL or BMDL for carcinogenicity shall be established.

II.6.3.7.1.5.

If carcinogenicity testing is not undertaken, the absence of such data shall be scientifically justified and the impact of its absence discussed.

II.6.4.1.1.

The need for safety data addressing other potential effects shall be determined on a case-by-case basis. VICH GL33 addresses the need for additional testing.

II.6.4.1.2.

Factors to be taken into account when considering the need for such data include:

II.6.4.2.1.1.

If relevant effects are seen in repeated dose or other toxicity studies (e.g. changes in lymphoid organ weights and/or histology and changes in cellularity of lymphoid tissues, bone marrow or peripheral leukocytes) additional functional testing may be required. The investigator shall justify the nature of any additional testing, taking account of the observations noted in other toxicity studies.

II.6.4.2.1.2.

For certain classes of substance (such as beta lactam antibiotics) which are known to elicit hypersensitivity (allergic) reactions in sensitive individuals data shall be provided on exposure levels that have been associated with hypersensitivity responses.

II.6.4.2.1.3.

Details shall be provided of all immunological studies performed with the substance as part of any aspect of the assessment (e.g. sensitisation assays performed for user safety or efficacy studies performed on immune-modulatory substances). Any reports of adverse effects in humans shall also be provided.

II.6.4.2.1.4.

Data obtained from such studies shall be taken into account when determining the toxicological ADI or alternative limit.

II.6.4.2.2.1.

Neurotoxicity testing shall be required where repeated dose studies indicate that there may be a relevant concern.

II.6.4.2.2.2.

Substances that have been shown in other toxicological assays to cause histological, biophysical or biochemical changes to the nervous system, or to cause neuro-behavioural changes, shall also be tested for neurotoxicity. Physicochemical properties, structure-activity information and recorded adverse effects in humans may give further indication on the need for neurotoxicity tests.

II.6.4.2.2.3.

Neurotoxicity testing shall be performed using the oral route and shall follow the advice given in the Organisation for Economic Cooperation and Development (‘OECD’) Guidelines for the Testing of Chemicals — Test Guideline 424 ( 17 ) on the methodology to be used in neurotoxicity studies in rodents. This study may be performed as a stand-alone study or may be incorporated in other repeated dose toxicity studies.

II.6.4.2.2.4.

Although OECD Test Guideline 424 does not specifically address effects on the activity of acetylcholinesterase, this end point shall be included in all repeated dose toxicity studies for specific substances known or suspected to have such activity (for example, organophosphates or carbamates). Testing for cholinesterase inhibition shall at least include measurements in brain and erythrocytes.

II.6.4.2.2.5.

If a substance has been shown to cause neuropathology or neurotoxicity in adults, or cause other types of toxicity indicative of nervous system involvement at a developmental stage, developmental neurotoxicity testing may be considered necessary. In such a case, OECD Test Guideline 426 ( 18 ) which advises on the methodology to be used in developmental neurotoxicity studies shall be followed. The extended one generation reproductive toxicity study (OECD Test Guideline 443 ( 19 )) also provides for developmental neurotoxicity testing.

II.6.4.2.2.6.

Organophosphates shall be tested for delayed neurotoxicity in a hen assay that incorporates measurement of neuropathy target esterase (‘NTE’) in brain tissue. Both single exposure (OECD Test Guideline 418 ( 20 )) and repeated exposure (OECD Test Guideline 419 ( 21 )) shall be considered. While single dose studies performed according to OECD Test Guideline 418 may only allow identification of a delayed neurotoxicity effect, repeated dose studies (OECD Test Guideline 419) may allow identification of a NO(A)EL or BMDL.

II.6.4.2.2.7.

The neurotoxicity studies shall allow the establishment of NO(A)ELs or LO(A)ELs or BMDL which shall be taken into account when determining the toxicological ADI or alternative limit.

II.6.4.3.1.1.

For substances with antimicrobial activity, antimicrobial effects on the human intestinal flora may occur at doses below those seen to induce toxicity in the toxicity tests. For such substances, a microbiological ADI shall be established in line with VICH GL36: Studies to evaluate the safety of residues of veterinary drugs in human food: general approach to establish a microbiological ADI  ( 22 ).

II.6.4.3.1.2.

The data shall be used to derive a microbiological ADI.

II.6.4.3.1.3.

The risks that result from residues shall be clearly distinguished from the potential risk to public health associated with the ingestion of food of animal origin which contains resistant bacteria selected under the pressure of an antimicrobial therapy.

II.6.4.3.1.4.

As described in the VICH GL36, the following two endpoints of concern shall be addressed in relation to the establishment of a microbiological ADI:

II.6.4.3.1.5.

Any departures from the established guidance shall be justified and the impact discussed.

II.6.4.3.1.6.

If no testing for effects on the human intestinal flora is undertaken, the absence of such data shall be scientifically justified and the impact of its absence discussed.

II.6.4.4.1.

Any available data on health effects seen in humans following exposure to the substance shall be provided. Such data may relate to intentional exposure of humans (e.g. when the substance is used in human medicine) or unintentional exposure (e.g. reports of occupational exposure). Such data may focus on epidemiological, pharmacological, toxicological or clinical findings.

II.6.4.4.2.

The data related to exposure of humans may provide valuable additional information on the overall toxicological profile of the substance, as well as provide information on the comparative sensitivity of humans and animals, even if they cannot be used for derivation of the ADI. In some cases such data may be useful in supporting arguments relating to the relevance (or lack of relevance) of certain findings in laboratory animals.

II.6.5.1.

If relevant safety evaluations of the substance have been undertaken by other EU or international scientific bodies including EFSA, ECHA, JECFA and FAO/WHO Joint Meetings on Pesticide Residues (‘JMPR’) this shall be highlighted, along with the conclusions reached.

II.6.6.1.1.1.

The toxicological ADI shall be derived by dividing the selected toxicological NO(A)EL/BMDL by an uncertainty factor, in order to take account of possible inter-species variation (i.e. differences in sensitivity of humans and laboratory animals) and intra-species variation (i.e. differences in sensitivity within the human population). The uncertainty factor may be adjusted to take other uncertainties into consideration, as necessary (see below).

II.6.6.1.1.2.

The formula used to determine the toxicological ADI shall be as follows:

ADI (mg/kg bw/day) = NOAEL or BMDL (mg/kg bw/day) divided by Uncertainty Factor

II.6.6.1.1.3.

The choice of the NO(A)EL or BMDL and the uncertainty factor shall be justified.

II.6.6.1.1.4.

Unless otherwise justified, the toxicological ADI shall be derived from the lowest NO(A)EL or BMDL observed in the most sensitive species in the toxicology studies. In certain circumstances a justification for using an alternative starting point may be possible (for example, if data exist demonstrating that the effect seen at LO(A)EL in the most sensitive species is not relevant for humans).

II.6.6.1.1.5.

If using the benchmark dose (‘BMD’) approach, the BMDL shall be used as the point of departure for derivation of the ADI. In most cases the choice of the critical endpoint is not expected to change when using the BMDL versus the NO(A)EL approach, since the same biological considerations apply.

II.6.6.1.1.6.

In selecting the default values for the magnitude of the response for which the BMDL is derived (i.e. the benchmark response (‘BMR’)), on choice of the recommended dose-response models as well as on reporting the results of a BMD analysis, guidance found in the EFSA Scientific Opinion on Use of the benchmark dose approach in risk assessment ( 23 ) shall be followed.

II.6.6.1.1.7.

In relation to uncertainty factors, the default assumption is that human beings may be up to 10 times more sensitive than the test animal species and that the difference in sensitivity within the human population is a tenfold range. Therefore, assuming appropriate studies are available, an uncertainty factor of 100 shall usually be applied.

II.6.6.1.1.8.

Where the results of animal studies indicate teratogenic effects at doses that do not cause maternal toxicity, an overall uncertainty factor of up to 1 000 shall be applied to the NO(A)EL or BMDL for teratogenicity. For non-genotoxic threshold carcinogens an uncertainty factor of up to 1 000 may be used, depending on the mechanism involved.

II.6.6.1.1.9.

It may occur that the most sensitive endpoint is observed in a species and/or study where all dose groups produce significant effects compared to the control group. In such cases the BMDL approach shall be recommended to establish the point of departure (‘POD’) from which to derive an ADI. Alternatively, if the effect observed in the lowest dose is a sufficiently minor response, it may be possible to establish an ADI based on this LO(A)EL. In this case an additional uncertainty factor of 2 to 5 shall be used to take into account that the LO(A)EL reference point is an unknown distance above the ‘true’ threshold.

II.6.6.1.1.10.

The choice of uncertainty factors for use in deriving the ADI shall not depend on whether a NO(A)EL or a BMDL is taken as the POD.

II.6.6.1.1.11.

Where the ADI is to be set on the basis of human data, there is no uncertainty factor to be applied for extrapolation from animals to humans. Thus, when using good quality human data from which to derive an ADI, it is appropriate to apply an uncertainty factor of only 10, to account for variation in individual responses between human beings.

II.6.6.1.1.12.

The refinement of the standard approach for selecting uncertainty factors may be acceptable where adequate justification is provided. For example, the use of (metabolic) pathway related uncertainty factors may be appropriate to refine the standard uncertainty factor used for inter-individual (intra-species) variability.

II.6.6.1.1.13.

Further refinement of the intra-species and inter-species tenfold uncertainty factors may be possible on a case-by-case basis, when toxicokinetic and toxicodynamic data support such adjustment factors.

II.6.6.1.1.14.

For the multiplication of uncertainty factors the use of probabilistic approaches may be appropriate.

II.6.6.1.1.15.

The use of these and other approaches for the refinement of standard uncertainty factors shall be fully justified.

II.6.6.1.1.16.

Having regard to the previous considerations, the uncertainty factor used shall usually have a value between 10 and 1 000 . Other values may be considered with appropriate justification.

II.6.6.1.2.1.

Pharmacological ADIs shall not be derived for all pharmacologically active substances as relevant pharmacological endpoints may be included in the toxicology studies. In such cases separate toxicological and pharmacological ADIs may not be needed.

II.6.6.1.2.2.

Guidance on the need for a pharmacological ADI as provided for in the Committee for Medicinal Products for Veterinary Use (‘CVMP’) guideline on the approach to establish a pharmacological ADI ( 24 ) shall be followed. Where no pharmacological ADI is derived, justification for its absence shall be provided.

II.6.6.1.2.3.

Where a pharmacological ADI is needed, the approach for its derivation shall be analogous to that described above under Section II.6.6.1.1 in relation to derivation of the toxicological ADI. The only difference is that the starting point used for derivation of the pharmacological ADI shall be the lowest NOEL or BMDL observed in the most sensitive species in the pharmacology studies.

II.6.6.1.3.1.

As described in Section II.6.4.3 microbiological ADIs shall be derived for substances with antimicrobial activity. The methodologies for establishing a microbiological ADI are detailed in VICH GL 36 and shall be followed.

II.6.6.2.1.1.

For most minerals and trace elements there is a natural baseline level in human body compartments resulting from their uptake from food and other environmental sources, and element specific homeostatic or accumulation processes. It is important to discriminate between essential trace elements for which there is both a minimum daily dietary requirement and an upper acceptable intake level and non-essential elements which are considered as undesired or even toxic for humans.

II.6.6.2.1.2.

The ADI approach is not appropriate for use in the assessment of essential elements as effects may occur at very low exposure levels, which represent a deficiency in supply. For most minerals and trace elements recommended dietary intake levels have been established by relevant scientific bodies (e.g. EU/EFSA; WHO). Daily dietary exposure estimates for essential elements may be compared with appropriate reference values, such as the recommended daily intake (‘RDI’), dietary reference values (‘DRVs’, previously: recommended daily allowances (‘RDA’)), tolerable daily intakes (‘TDIs’) or tolerable weekly intakes (‘TWIs’) and provisional tolerable weekly intakes (‘PTWI’). These values may be used in the risk assessment, in a way analogous to the ADI. The combined exposure resulting from treatment related residues and exposure from dietary and natural sources shall not exceed the respective reference values.

II.6.6.2.1.3.

This approach may be appropriate for minerals, elements, vitamins and other natural constituents of food for which relevant recommended dietary intake levels have been established.

II.6.6.2.2.1.

When consumer exposure to residues of the active substance in food of animal origin is negligible or very low compared to the level of exposure that already occurs as a result of the presence of the material in the environment or already present in commodities (most notably foods), then it may be possible to argue that the impact (in terms of consumer exposure to residues) of the proposed use in veterinary medicinal products is negligible and that the establishment of an ADI shall not be necessary. A worst case estimate of residue levels that may occur as a result of the proposed use of the substance shall be provided, along with an estimate of the resulting consumer exposure. This shall be compared with the level of exposure known to occur via other sources. This approach may be particularly relevant for herbal medicines and vegetable extracts, as well as for natural organic acids (e.g. oxalic acid).

II.6.6.2.2.2.

The chemical make-up of herbal/vegetable based products (including extracts) is typically complex and may be quite different to the make-up of residues that will remain in food commodities derived from treated animals. Due to the complexity of the parent material, it may not be practical or even possible to identify the resulting residues. For such substances, an alternative to the standard ADI based approach may be appropriate.

II.6.6.2.2.3.

When using this approach it is important to exclude any possibility of non-threshold effects such as genotoxicity.

II.6.6.2.3.1.

If the pharmacologically active substance is identical to an endogenously produced molecule, it may be possible to demonstrate that the level of consumer exposure that occurs as a result of residues in food of animal origin is insignificant compared to the level of human exposure to the endogenous substance.

II.6.6.2.3.2.

Human exposure to such substances may be expected to come from both exogenous (treatment related residues plus natural levels in food of animal origin) and endogenous (human physiological) origin. The risk assessment of the residues is complicated by the difficulty in assessing the likely response of ingestion of low exogenous levels when humans are constantly exposed to relatively high and fluctuating levels of endogenously produced substance and fluctuating dietary levels. In addition, for many active substances (like hormones, corticosteroids) exogenous exposure may lead to regulation of the endogenous production that, in turn, may change the endogenous hormone levels and the overall response. This complicates the interpretation of conventional toxicology studies and the derivation of an ADI. In addition, findings in laboratory animals may be difficult to extrapolate to the situation in humans owing to complex specific differences in biochemical/pharmacodynamic regulatory mechanisms.

II.6.6.2.3.3.

Consumer exposure to residues may be best estimated by comparing treatment related excess intake of residues from food to intake of the substance from untreated animals (with natural background levels). This may then also be compared with the endogenous daily human production of the substance. Possible species specific differences (analogues) shall be discussed.

II.6.6.2.3.4.

This approach may be appropriate for hormones and other endogenously produced substances.

II.6.6.2.4.1.

For substances that are not absorbed following oral ingestion, systemic exposure is negligible (or even non-existent). For such substances it is not possible to establish a conventional oral NO(A)EL or BMDL and ADI. The risk assessment for these types of substances shall normally rely on demonstration of the absence of oral bioavailability in suitable models or, where appropriate, through proof of degradation and/or inactivation under gastric conditions (likely to be demonstrated in in vitro models). In addition, for such substances, possible local effects on the gastrointestinal system (including microbiological effects on the colonisation barrier) shall be addressed.

III.2.1.

A detailed and critical summary of the residues file shall be required for all applications.

III.2.2.

The detailed and critical summary shall:

III.2.3.

Annexes to the detailed and critical summary shall include:

III.3.1.

Metabolism and residues data shall be required to characterise residues present in relevant food commodities, to demonstrate the time course of their depletion to a safe level (usually based on the ADI) and so to allow derivation of MRLs.

III.3.2.

The data shall be provided in the form of a total residues depletion study providing quantitative data on the parent drug and its major metabolites in relevant food commodities, and the change in the levels of these over time. Total residues studies usually use radiolabelled drug although data from non-radiolabelled studies may be provided where appropriate (for example if the substance is known not to be metabolised). A separate marker residue depletion study shall often also be provided, using unlabelled drug and monitoring the depletion of the marker residue in relevant food commodities over time. Total residues and marker residue data may be provided by means of a single radiolabelled study that also uses an appropriately validated non-radiolabelled method to monitor depletion of the marker residue.

III.3.3.

The test material shall contain the substance of concern in a representative concentration. It shall be administered by the intended route of administration of the proposed product, at the highest intended dose and for the maximum intended duration of treatment or for the time required for steady state to be achieved in edible tissues. Studies shall be conducted in animals that are representative of the proposed target populations.

III.3.4.

Guidance provided in VICH GL46: Studies to evaluate the metabolism and residue kinetics of veterinary drugs in food-producing animals: metabolism study to determine the quantity and identify the nature of residues  ( 25 ) shall be followed in order to monitor (quantify) the depletion of total residues and key metabolites over time. These studies shall normally be performed using radiolabelled drug.

III.3.5.

Guidance provided in VICH GL49: Studies to evaluate the metabolism and residue kinetics of veterinary drugs in food-producing animals: validation of analytical methods used in residue depletion studies  ( 26 ) shall be followed in order to demonstrate the analytical method standards and in order to obtain marker residue depletion data of an acceptable quality.

III.3.6.

Specific guidance relating to residue studies to be undertaken for substances intended for use in honey bees provided in VICH GL56: Studies to evaluate the metabolism and residue kinetics of veterinary drugs in food-producing species: study design recommendations for residue studies in honey for establishing MRLs and withdrawal periods  ( 27 ) shall be followed.

III.3.7.

The total residues study (usually performed with radiolabelled drug) shall provide information on:

These data shall be used to establish the marker residue and the ratio of marker to total residues for each relevant food commodity.

III.3.8.

A suitable marker residue shall be identified. The marker residue may be the parent drug, any of its metabolites or a combination of any of these. The marker residue shall have the following properties:

III.3.9.

The ratio of marker to total residues describes the relationship between the marker residue and total residues in each relevant food commodity. This ratio may be different in different food commodities and, as it may vary over time, it shall be established until the time corresponding to that at which residues of concern are expected to be below the ADI. The ratio of marker to total residues shall be used in the intake calculation to calculate potential consumer exposure to total residues from data relating to the marker residue.

III.3.10.

By monitoring the depletion of total residues in the edible tissues/food commodities, the time point at which total residues deplete to below the ADI (or the fraction of the ADI available for use) shall be established. In each tissue/food commodity, the concentration of the selected marker residue at this time point shall be taken as the starting point from which the MRL shall be developed.

III.3.11.

Information from the metabolism study shall also allow comparison of the metabolites produced in the target animal species with those produced in the laboratory animals species in order to ensure that the major residues to which consumers will be exposed (i.e. the major metabolites produced in the target species) were adequately tested in the laboratory animal toxicity studies.

III.3.12.

Any departures from established guidance shall be justified and the impact discussed.

III.4.1.

Monitoring or exposure data of the pharmacologically active substance shall not be required. However, if available, it may provide valuable additional information in certain cases, i.e. for substances that are already present in the environment (either naturally or as a result of use in the veterinary or other sectors). Such data may be useful in determining background levels to which consumers may already be exposed. If such data are available, whether as published results from official residue monitoring bodies or as results of academic or other research, these shall be provided.

III.5.1.

A validation report of the analytical method used for quantification of the marker residue in the residues study shall be provided. Validation shall demonstrate that the analytical method complies with the criteria applicable for the relevant performance characteristics. The specific guidance on validation of analytical methods is provided in VICH GL49 shall be followed.

III.5.2.

Analytical methods shall be provided at least for those food commodities and species in which MRLs are requested.

III.5.3.

The availability of standards shall be confirmed and contact details provided in order to allow an exchange of information, if necessary, between representatives of the EU and national reference laboratory staff and the company.

III.5.4.

Any departures from the requirements above shall be justified and the impact discussed.

III.5.5.

The analytical method shall be evaluated for compliance with VICH GL49 and the additional points raised above. In addition, the Agency shall consult the European Reference Laboratory for control of residues for the particular substance type on the adequacy of the available methods and validation data.

III.5.6.

Following the Agency's opinion, the validation data may be shared with other EU and national reference laboratories in order to facilitate development of appropriate methods by those authorities.

III.6.1.

The residues evaluation shall include an assessment of the potential effects of microbiologically active residues on microorganisms used for industrial food processing, in particular as regards the manufacture of dairy products.

III.6.2.

The data shall be used to establish a residue concentration without effect on starter cultures. This shall be taken into consideration when deriving MRLs, to ensure that residues present in relevant food commodities (i.e. milk) are not present at levels that impact on dairy starter cultures.

III.6.3.

The studies to be performed shall follow Agency's guidance for the assessment of the effect of antimicrobial substances on dairy starter cultures ( 28 ).

III.6.4.

Any departures from the established guidance shall be justified and the impact discussed.

III.6.5.

If no testing of microorganisms used for industrial food processing is undertaken, the absence of such data shall be scientifically justified and the impact of its absence discussed.

III.7.1.

If relevant residues evaluations of the substance have been undertaken by other EU or international scientific bodies including EFSA, ECHA, JECFA and JMPR this shall be presented, along with the conclusions reached.