Calling all animal researchers & toxicologists - reliability question

Doug K dkochel at icdc.com
Mon Mar 8 15:30:11 EST 1999

A simple question, huh?

Obviously, far more testing is required to satisfy your query. What kind of
study and how many animals do you think it would take to verify this
correlation? Think of the toxicity testing, how many substances will kill or
sicken commonly used lab animals and yield same results when given to
humans? A well equipped lab could kill or sicken millions of animals in the
pursuit of the answer. Isn't it at least a little better to tackle this one
substance at a time as the need arises?

Perhaps a simpler question would be: Why do the researchers need to produce
evidence that falsifies a claim made by a group of individuals who wish to
make their job more difficult? We're back to the "burden-of-proof" argument
then. I haven't seen any AR types win that one yet.

Doug K.
"Ride well, shoot straight, speak the truth"

greenrd at hotmail.com wrote in message <7c13v8$fa0$1 at nnrp1.dejanews.com>...
>I have a simple question. The reason I ask is because some
>have gone so far as to claim that "no drug safety or efficacy results can
>reliably extrapolated from any species to any other species".
>My question: Where are the published, peer-reviewed studies that falsify
>To be more precise: Can anyone supply a references for repeated studies
>demonstrate a reliable correlation between the toxicity and/or efficacy of
>drugs used to treat several laboratory-induced diseases in one species,
>the toxicity and/or efficacy of those same drugs used to treat
>"naturally-induced" (i.e. not laboratory-induced) diseases in another?
>Several obvious points should be apparent: Post-hoc reviews of previous
>studies are likely to be invalid, because they would have to ignore
>unpublished studies - besides which, "approximately 1% of papers published
>medical journals are scientifically sound", so they would have to be very
>careful about which ones they chose. Testing only one drug is far too
>selective - what is really at issue here is whether species-extrapolation
>scientifically valid as a general deductive step, not whether there are
>correlations in the cases of particular drugs. Testing only drugs that are
>already known to demonstrate a correlation would also be selective, not to
>mention pointless since the results are a foregone conclusion (that's
>assuming that the original trials were sound, which is doubtful - see
>quotation above). Examining only laboratory-induced diseases in both
>would do nothing to falsify another key claim of the antivivisectionists:
>that laboratory-induced diseases are *inherently* unscientific models of
>"naturally-induced" diseases. (Well, they might have to make an exception
>radiation poisoning, say, which in *one* sense of the word is
>"naturally-induced" - in another sense it is highly unlikely to be, being
>caused by nuclear bombs etc.)
>I suspect there are no such studies, and I suspect this is because those
>antivivisectionists are absolutely right. If there are no such studies,
>tell me, on what scientific evidence is species-extrapolation based? Please
>don't fob me off with talk of "anatomical and physiological similarities" -
>the scientific route is clear - in order to prove that it is reliable to
>extrapolate in such a manner, one must demonstrate a reasonably high
>correlation, by means of repeated, well-controlled, blind, unbiased,
>peer-reviewed studies.
>One might argue that we don't even require proof that an extrapolation from
>*human* clinical trial groups to the potential market of a human drug, a
>market which may span much of the world population geographically, is
>To which I say, indeed, good point. That children and adults often react
>differently to treatments is well-known - what other variations are there
>that we don't know about, or choose to ignore for the sake of profit?
>Robin Green
>-----------== Posted via Deja News, The Discussion Network ==----------
>http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own

More information about the Toxicol mailing list

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