Species-specific default factor

In the following text, various studies will be described, which attempt to establish a scientific rationale for the selection of the interspecies assessment factor. Based on these studies, it can be concluded that a species-specific default factor based on differences in caloric requirement (see Table 5.4) should be used for interspecies extrapolation regarding metabolic size. The remaining interspecies differences should preferentially be described probabilistically, or a deterministic default factor of 2.5 could be used for extrapolation of data from rat studies to the human situation. 

ECETOC (2003) recommended that in the absence of any substance- or species-specific mechanism or PBPK modeling (Section 4.3.6), allometric seating based on metabolic rate (W° ) (caloric requirement approach. Section 5.3.2.3) is considered to provide an appropriate default for an assessment factor for interspecies differences with respect to systemic effects. Allometric scaling was stated as being a tool for estimating interspecies differences of internal exposure or body burden and to provide indirectly information on differences in sensitivity between species. Typical scaling factors for interspecies adjustment were noted as 7 for mouse, 4 for rat, and 2 for dog however. 

Ideally, the interspecies extrapolation should be based on substance-specific information however, for most substances, only limited or no data are available. Therefore, an assessment factor is usually applied in the interspecies extrapolation. The traditionally used default interspecies assessment factor is 10, possibly divided into a sub-factor of 4 for differences in toxicokinetics and a sub-factor of 2.5 for differences in toxicodynamics as proposed by Renwick (1993) and adopted by the WHO/IPCS (1994). The validity of the interspecies default factor of 4 for toxicokinetics has been assessed by Walton et al. (2001a,b) for each of the test species (dog, rabbit, rat, and mouse) the authors concluded that their assessment supports the need to replace the generic default factor by a compound-related value derived from specific, relevant, quantitative data. 

---