Day 038 - 19 Oct 94 - Page 16

     
     1
     2        This number is not a scientific result, but I think of it
     3        as an administrative convenience, lacking scientific
     4        legitimacy.  Nominally the assumptions underlying it are
     5        that humans might be 10 times more sensitive than
     6        laboratory animals, and that within the human population
     7        the variability within the human population might be
     8        captured by a factor of 10.  I find neither of those
     9        assumptions plausible.
    10
    11        On the contrary, it seems to me, where we do have evidence
    12        (and there is not enough of it), but where we do have
    13        evidence of the effects that chemicals have both on humans
    14        and laboratory animals, quite frequently that evidence
    15        indicates that the difference between humans and rats or
    16        mice or guinea pigs is substantially greater than to be
    17        captured with a factor of 10; and, moreover, it seems to me
    18        quite clear that the variability within the human
    19        population is also very much greater than the factor of
    20        10.  Sensitivity is a variable.  This is especially true
         for intolerance effects.  There are people who will
    22        exhibit, say, asthmatic symptoms or eczema or even
    23        epileptic symptoms on a hundredth or a thousandth of the
    24        level of exposure which can be readily tolerated by many
    25        other people.
    26
    27        So, if one is to invoke some scaling factor, or safety
    28        factor, to generate an acceptable daily intake by reference
    29        to the results of the animal tests, I would think that
    30        figure ought to be substantially higher than a 100; several
    31        orders of magnitude higher, perhaps.
    32
    33   MS. STEEL:   Can you explain why it is not reasonable to assume
    34        that almost all toxic effects, other than genotoxic
    35        carcinogenesis, are governed by some identifiable
    36        threshold?
    37        A.  Yes.  The threshold assumption is very important to
    38        toxicology policy making, but there is precious little
    39        evidence to indicate that it is a sound assumption.
    40        Perhaps it would help if I just dealt with genotoxic
         carcinogens first, because that is the clearest case where
    42         -- I was going to say "nobody"; perhaps I had better
    43        qualify that and say "almost" nobody claims that one can
    44        find a threshold.  There are a small number of people who
    45        seek to make that case, but it is not a case that is
    46        generally accepted.
    47
    48        So, genotoxic carcinogens are amongst the most hazardous
    49        group of compounds, at least as far as long-term toxicity
    50        is concerned.  These are compounds which, as far as we 
    51        understand it, can initiate or promote tumour, malignant 
    52        tumours, and the mechanism whereby they do that involves 
    53        damaging the genetic material in the nucleus of the cell.
    54
    55        These compounds are so powerful that one molecule by itself
    56        on its own might be sufficient to trigger that genetic
    57        change in the nucleus of the cell and thereby might be able
    58        to initiate the development of a tumour.  Clearly for
    59        genotoxic carcinogens, I think, for the purposes of policy
    60        making, it has to be assumed that there is no safe level.