(Paul Whitehead, DABT) writes:
> Lawrence Segal (am879 at FreeNet.Carleton.CA) wrote:
> :
> : I am currently reviewing information concerning a proposed occupational
> : exposure limit for an industrial chemical. The substance is a
> : non-genotoxic carcinogen, which has been tested in both mouse and rat
> : chronic/onco studies. In males of both species, statistically significant
> : increases in liver adenomas/carcinomas were observed at exposure levels of
> : 75 ppm. The company that manufactures the substance is proposing a TWA
> : occupational exposure limit of 3 ppm. This provides a 25-fold
> : uncertainty factor applied to the LOEL in the rodent studies. The company
> : considers this to be an adequate safety factor, because the liver tumours
> : were benign, there was no significant increase in mortality at 75 ppm
> : relative to control, and the latency period was long.
>> -I think that an adequate safety margin could depend on the mechanism
> of action of the non-genotoxic carcinogen, eg is it a peroxisomal
> proliferator? If this is the case, then there is a considerable body
> of evidence to show that such a chemical would not be expected to be a
> carcinogen in man.
As I recall the data summary, the substance is not a peroxisome proliferator.
> A safety margin could then be set based on inter
> and intra species sensitivities (eg 10x for each based on animal NOEL)
> plus any additional factors that could modify this based on
> metabolism, etc.
This is in line with what I would have considered appropriate for this
substance and data.
>> If the mechanism is unknown, and you are basing your comment re
> non-genotoxicity on the fact that it is negative in an appropriate
> (and validated) battery of STT's, then one must conclude that the
> chemical could possibly be carcinogenic in man under appropriate
> exposure conditions.
It was well studied for genotoxicity and was negative in all assays.
> However, non-genotoxic carcinogens are considered
> to have a threshold. In any this case I believe a 25x factor is too
> low, and as previously mentioned, the default figure is usually 100x,
> but with additional safeguards to take into account the uncertainly
> over the mechanism of action, eg an additional 10x.
I agree with your assessment.
> Paul Whitehead BSc DABT
--
Lawrence Segal
Ottawa, Ontario, CANADA
