280 EFFECTS OF CHEMICALS
included as system parameters in evaluation of pollutant
effects. It is possible that human tests may become necessary,
but moral and legal considerations pose serious questions.
Agents under test must be administered at subtoxic,
toxic, and chronic levels. Effects of possibly significant but
normally non-toxic agents on the system containing the toxic
agent under investigation must also be evaluated.
Testing procedures must be realistic and reflect the path-
ways of human exposure. However, the exception to this
rule is in sensitivity. Evaluation of carcinogenic effect may
require administration to test animals by different means.
Normal ingestion in humans might be by inhalation while
adequate dosage in rats might require addition of the sub-
stance in question to daily food. This is particularly true
when dealing with weak environmental carcinogens.
Metabolic compatibility between a test species and
humans is important for extrapolations of test results. There
is seldom a one to one correspondence and most authori-
ties feel that at least two test species should be examined.
There is much testing data on rodents, pigs, and subhuman
primates. Thus, further utilization of these animals is indi-
cated. In special cases, however, a less common species may
be necessary in order to gain the desired data.
It is necessary to test levels much higher than human
exposure for carcinogenic, teratogenic, and mutagenic sub-
stances. It must be noted that, even when large numbers of
mice are tested, the number is still small compared to the
millions of humans that might be exposed in the everyday
environment. Not all humans exposed would respond, in any
event, to the dangerous substance. As an example, meclizine,
an antihistamine used for treatment of morning sickness, has
been found to be teratogenic in the rat but not so in a relatively
small number of women tested. The question as to effects on
a larger population is unanswered. For thalidomide, humans
are found to be 60 times more sensitive than mice, 100 times
more sensitive than rats, 200 times more sensitive than dogs,
and 700 times more sensitive than hamsters. It is obvious,
therefore, that predictions as to teratogenic doses of thalido-
mide on the basis of animal testing would mean very high
and dangerous exposure levels.
Complicating the testing picture is the increased effect of
substances when a second, supposedly innocuous, substance
greatly increases the undesirable end effect. Such reactions
were the basis of the Delaney Amendment of 1968 to the Food,
Drug, and Cosmetic Act (US). The Amendment required that
no additive be considered safe if found, under appropriate
testing conditions, to induce cancer in man or animal.
Recent advances in molecular biology have made it pos-
sible to give weight to a broad range of evidence, including
details as how toxic agents affect human cells and on genetic
material which controls cell reproduction. The relatively
new concept of virtual risk may find application. It should
be noted, however, that use of cost/benefit may not be desir-
able and should be treated with caution, if not suspicion. At
times this concept has been used to evade existing environ-
mental regulations. Elected officials are finding that efforts
to weaken public health and pollution control legislation are
unpopular.Often, non-carcinogenic alternatives are available as
replacements for questionable materials. There is consider-
able controversy about the pesticide DDT. Some authorities
question the continued utility of DDT for control of cotton
insects due to development of resistant strains. This contro-
versy will not soon die down. Cyclamates, banned in recent
years, were an intentional food additive and of no value
nutritionally. Only after their widespread use for a number
of years was there shown a carcinogenic danger.
The argument of economic loss to special interest groups
is of no validity. In view of the uncertainty of long-term
effects of carcinogenic materials, zero tolerance levels must
be imposed for these substances.
It is important that chemicals, and their derivatives,
suspected or implicated as toxic, carcinogenic, teratogenic,
and mutagenic be detected and monitored in the environ-
ment. Epidemiological studies may show effects but it is
necessary to quantitively establish occurrence of these sub-
stances. It has been possible to demonstrate the relationship
between cigarette smoking and lung cancer although some
special interest groups seriously disputed this. However,
in this case were two dissimilar populations, smokers and
non-smokers. In the case of exposure of the general popu-
lation to a suspect substance, evaluation and epidemiologi-
cal treatment of data are more difficult. Apart from rubella
virus, no known teratogens, such as ionizing radiation,
mercury, etc. have been positively identified epidemiologi-
cally in the highly industrialized countries. This indicates
the great need for better and more comprehensive gather-
ing of data relating to birth defects. It is to be hoped that
environmental effects should be segregated from natural
and spontaneous mutation effects. It would be necessary to
monitor special indicator traits. Such traits would have to
be chosen with great care and followed for a considerable
period.
Recently it has been suggested that mutation rates
could be monitored by means of data on spontaneous abor-
tions. Teratogens may cause chromasome aberrations but
these act after fertilization and would be against a normal
background.
The question of data on occupational hazards, long the
major field of endeavor of the industrial hygienist, must be
raised in connection with legislation thought to be inade-
quate and outdated. The correctness of much of the informa-
tion on which maximum allowable concentrations are based
is in doubt at this time. Probably a critical reexamination
will soon come about as a result of increased environmental
awareness.
Present toxicological techniques are not sufficiently
sensitive to monitor adequately many weak carcinogens.
Compounding the problem is the difficulty of recognizing
effects of many substances in the general population. Special
situations as to sample population at times help to simplify
the problem, such as cigarette smoking. But there is cer-
tainly a great area in which problems are strongly suspected
to exist but statistical validation is not possible at this time.
Yet tolerance and allowable concentration limits must be set.
There is a most fruitful area for research here.C005_004_r03.indd 280C005_004_r03.indd 280 11/18/2005 10:21:39 AM11/18/2005 10:21:39 AM