Page 152 - Veterinary Toxicology, Basic and Clinical Principles, 3rd Edition

Regulatory Aspects for the Drugs and Chemicals Used in Food-Producing Animals Chapter | 7 119

VetBooks.ir the starting point (WHO, 1989). In order to arrive at such departure (POD) from background level response through
the use of the benchmark dose (BMD) as a statistical esti-
NOAEL, a battery of standardized toxicity tests are carried
out (i.e., acute toxicity, genotoxicity studies (mutagenicity,
mate of the NOAEL. Health-based guidance values
clastogenicity), subchronic oral toxicity, chronic oral (HBGVs) (i.e., an ADI for a chronic intake or an ARfD
toxicity/carcinogenicity, and specific studies concerning for an acute intake) are most often established using a
reproduction and developmental toxicity among others) POD from a toxicity study in experimental animals. The
(EC, 2008b). The ADI is calculated by dividing this by a most frequently used POD is either the NOAEL (mg/kg
suitable safety or uncertainty factor, conventionally of bw per day) or the BMD, which is a dose producing a
100, to account for the differences between test animals low but measurable adverse response. The lower one-
and humans (factor of 10) and possible differences in sided confidence limit of the BMD (the BMDL) is the
sensitivity between humans (another factor of 10). POD for the derivation of a HBGV or for calculation of a
Depending on the available data set and the nature of margin of exposure (MOE). The MOE is derived by tak-
substance under study, higher or lower safety factors may ing the ratio of the POD for the most relevant, sensitive
be applied (e.g., small size of the animal test group, end-point to an estimate of exposure by a high consumer.
nature of the critical effects, quality of the data). This Where appropriate, acute dietary exposure estimates
concept cannot be applied in cases where certain geno- should cover a time period of food consumption over a
toxic carcinogens do not show a dose-dependent thresh- single meal or 24 h and are intended to be used for com-
old level. parison with ARfD values in a RA process. It is important
ADIs are related to chemical substances that are delib- to consider data of short-term or acute studies.
erately added to a product or ingredient or which can be The HBGVs for veterinary drugs are generally estab-
found on food following, for instance, treatment of crops lished. However, when this is not possible or not appro-
with pesticide sprays or antifungal agents. A Tolerable priate, the MOE could be calculated. There are primarily
Daily Intake (TDI) is an estimate of the quantity of a two situations when this might be done: (1) when the
chemical contaminant to which we may be exposed nature of the end-point is such that derivation of a HBGV
through environmental contamination, and which when is not appropriate, and (2) when there are deficiencies in
found in food can be ingested daily over a lifetime with- the database such that it is not possible to derive a HBGV
out posing a significant health risk. The TDI is used for with confidence.
contaminants, whereas the ADI is used in cases where The MOE approach is specially used for a number of
exposure can be controlled, such as feed additives and genotoxic carcinogens using BMDL values derived by
residues of pesticides and veterinary drugs in food. fitting a range of models to the available experimental
Guidance values may be expressed as a TDI, provisional dose response data. The MOE approach can also be
maximum tolerable daily intake, provisional tolerable used in characterizing risks associated with certain con-
weekly intake, or provisional tolerable monthly intake. taminants in food for which there are deficiencies in the
The ADI represents total drug residues, parent and all database and it is not possible to derive a HBGV with
metabolites, that can be safely consumed daily throughout reliance. In general, the acceptability of an MOE is based
one’s life. ADIs also can be developed from pharmacolog- on considerations similar to those used in the derivation
ical or microbiological data based on no-pharmacological of a HBGV. An MOE of at least 100 (10-fold for each of
observed effect levels. However, the impact of low levels interspecies and intraspecies variability) would be con-
of antibiotics on the intestinal microflora is not examined sidered acceptable for the effects with a biological
directly in these toxicology studies. For substances with threshold. For effects without a threshold, some regula-
microbiological activity what is actually used is the major tory bodies have suggested an MOE of greater than
adverse microbiological effect arising from the effects of 10,000. This is the case with genotoxic and carcinogenic
residues of antimicrobial drugs in food of animal origin compounds, where the use of the NOAEL is not appro-
acting on the human gastrointestinal bacteria flora, particu- priate. For such compounds, where a BMDL cannot
larly those acting on the colonic flora. This approach is determined, a T25 (the chronic daily dose in mg/kg bw
currently used by CVMP and JECFA (Joint FAO/WHO that will give 25% of the animals tumors at a specific tis-
Expert Committee on Food Additives) but has never been sue site, after correction for spontaneous incidence,
used by the FEEDAP Panel as can be seen in Commission within the standard lifespan of that species) is calculated
Regulation (EC) No. 429/2008 (EC, 2008b). (FAO/WHO, 2014).
The use of mathematical and statistical approaches for The BMD is based on modeling all of the available
hazard characterization appears to be valuable. One dose response data and generally estimates an excess
assessment approach that has been used in recent years is risk of 10% over background for the response of interest.
the use of statistical estimates based on all the available The benchmark dose lower limit (BMDL) uses the lower
data in a dose response series to determine the point of bound of a 95% confidence limit on the benchmark dose