Animal Models in Toxicology/JBudny post 17Feb96

George L. Wolff GWOLFF at NCTR.FDA.GOV
Wed Feb 21 10:56:01 EST 1996

 The point I tried to make in my previous post, obviously 
unsuccessfully, was that specific animal models are valid as human 
substitutes ONLY if the biochemical/physiological patterns which give 
rise to/result in the phenotypic endpoint of interest are either 
identical (most unlikely) or closely similar in both species. 
    Inborn errors due to similar single gene mutations in humans and 
animals would appear to be straightforward candidates for 
investigation in animal models; however, since the background strain 
genomes are different, the phenotypic expression of the inborn error 
may differ at the organismic level in the two species. Thus, even 
here, caution is necessary in extrapolating from animal model to Homo 
    Dose response types of toxic responses in animal models are, 
similarly, subject to the expression of gene patterns in the 
background strain genome which may differ between the animal model 
and humans. In fact, even within genetically homogeneous animal 
populations, there is demonstrable phenotypic variation with respect 
to tumor formation in response to the same dose level of test agent. 
I discuss this in extenso in the February issue of Fundamental & 
Applied Toxicology 29:176, 1996 under "Variability in gene 
expression and tumor formation within genetically homogeneous animal 
populations in bioassays." 
    Homo sapiens is genetically relatively heterogeneous in 
comparison with the genetically homogeneous F1 hybrid and inbred 
animal populations used in toxicological bioassays. Therefore, the 
obvious question is always to WHICH subgroup of men/women can the 
animal test data be extrapolated with any sort of validity?
    Since toxic agents cannot, ethically, be tested in a variety of 
human populations, what is the alternative to the use of test animals?
In vitro tests cannot provide information on the toxic response of 
the intact organism although the toxic response of particular cell 
types can be elucidated in such systems. However, the organismic 
extracellular microenvironment - which plays a major role in the 
toxic response of cells - is absent and thus extrapolation to the 
whole organism is still fraught with doubt.
    The validity of using animal models and which are the "best" 
models to use for particular investigations has been discussed for 
decades without satisfactory resolution. Hopefully, by continuing 
this discussion over a prolonged period of time with input from 
toxicologists and others with diverse viewpoints, it should be 
possible, eventually, to arrive at a rational solution to this 
serious dilemma. At least, it should be fun trying!  

Tel: (501) 543-7522
FAX: (501) 543-7635/7662
NCTR       HFT-140    
3900 NCTR Road
Jefferson, AR 72079-9502

More information about the Toxicol mailing list

Send comments to us at biosci-help [At] net.bio.net