Laboratory animal experiments, carcinogen

Laboratory animal experiments demonstrate that dose-response relationships exist between carcinogen exposure and adduct levels, suggesting the utility of these studies In human exposure assessments. In animals, DNA and protein adduct levels correlate with dose of PAHs (63-65). tobacco-smoke condensate (43.60.66). H-nltrosamlnes (67-70). aromatic amines (21) (Beland, F. A., unpublished data Poirier, M. C., unpublished data) and heterocyclic amines (72.73). 

An increase in cancer appearing in more than one species or strain of laboratory animals or in more than one experiment is considered sufficient evidence in animals. Data from a single experiment can also be considered sufficient animal evidence if there is a high incidence or an unusual type of tumor induced. However, a carcinogenic response in only one species, strain, or study is normally considered as only limited evidence in animals. 

No indications on mutagenicity or carcinogenicity have been found in laboratory animal studies. Subcutaneous injections in mice led to a small increase in fetal malformations, but experiments with oral exposure of mice over several generations did not show any effects of toxicity to reproduction. 

Chrysene has elicited skin tumors in mice following chronic (68-82 weeks) dermal exposure. Topical application of a chrysene solution in n-dodecane/decalin to the skin of mice produced a significant increase in the carcinogenic potency of chrysene compared with the use of decalin alone 26% and 63% of mice exhibited papillomas and carcinomas, respectively, at 49 weeks (Horton and Christian 1974). Because only one dose level was employed, no dose-response relationship can be inferred, and no solvent control was included. However, in other experiments decalin and n- dodecane have been shown to be noncarcinogenic in mice (Bingham and Falk 1969). An average dose of 1.2 mg/kg/day is the lowest dose of chrysene that has been found to elicit malignant tumors in laboratory animals. 

The carcinogenic activity of Be, whether administered in the form of the metal, alloys, or other organometallic compounds, has been confirmed in a number of experiments on laboratory animals (Leonard and Bernard 1993), as has been the case of cadmium (see Table 9.5). Epidemiological studies have not provided clear evidence of a carcinogenic hazard of Al (Leonard and Gerber... 

In an NAT experiment, similar to those for NNN and NNK, NAT was not carcinogenic at any of the three dose levels. lARC (1870) concluded that the TSNAs are the most abundant suspected carcinogens in tobacco smoke. It considered NNN and NNK to be proven carcinogens for laboratory animals. It considered the evidence limited for defining the carcinogenicity of NAB in laboratory animals and inadequate to label NAT as a carcinogen for laboratory animals. 

Because of the successful induction of cancer in a laboratory animal by Yamagiwa and Ichikawa (4361) and the discovery that several PAHs were tumorigenic when painted on the skin of laboratory animals (194, 797, 2078), the tumorigenicity of literally hundreds of PAHs (and structurally similar nitrogen analogs) and their alkyl derivatives was studied from 1932 to 1941. Many of the assertions made about the correlation between the laboratory findings and human experience were extremely farfetched and caused much confusion. This led to the request for Shear of the U.S. National Cancer Institute to attempt to return order to the field of carcinogenicity. The result was the classical description by Shear and Leiter (3627), a description whose pertinence is still valid. 

Group IV includes weak carcinogens, which have induced tumours in certain laboratory animals, however, in general the experiments with these substances have so far yielded controversial results. 

Animal experiments, in particular, have been useful in testing various hypotheses about the relative effectiveness of different kinds of dietary fiber and about possible mechanisms of fiber effects. Thus, studies of bulking action (9-10), of selective secondary bile acid binding (11-12), of increased intestinal transit times (13), of altered bacterial activities and secondary bile acid production (14-19), and selective binding of carcinogens (16-19) have been carried out in laboratory animals. Several related studies have been conducted in humans as well. Among the latter are investigations of the relationship between the bulk fiber content of diets and fecal bile acids or other steroids (4.20-21). 

Animal studies. In many cases, much of the information on the toxicity of a solvent will have been derived from experiments using laboratory animals (usually rats or mice). Such experiments may have involved exposure by inhalation or by the oral or dermal routes, with duration ranging from a single exposure to the lifespan of the animal. Experimental protocols are available to cover general toxic effects, as well as skin and eye irritation, skin sensitisation, reproductive effects and carcinogenicity. 

Other Chronic Effects. Those solvents which are able to dissolve fats may cause dermatitis if there is prolonged skin contact, and those who handle solvents frequently should bear this in mind. The potential carcinogenicity of solvents is a matter about which concern is sometimes expressed, as are their possible reproductive effects. There is some rather poor evidence from animal experiments which purports to show that trichloroethylene is carcinogenic, but this has never been confirmed so far as human exposure is concerned. Some solvents can be shown to have mutagenic properties in laboratory tests but, again, there is no evidence which would lead one to believe that any of the solvents in common use can cause cancer. 

Of course, many scientists who worked in the laboratory setting were not so sure the results of their work should be taken quite so seriously. They doubted that results from high-dose studies in animals had much relevance to typical human exposures - such experiments would, at best, reveal some of the secrets of the ways carcinogens worked, but were of little use in predicting human risks. 

We also might be accused of creating a highly artificial situation, because the cells of genetically homogeneous animals held under strict laboratory controls do not experience nearly the number and type of host and environmental influences experienced by people exposed to the same chemical. Perhaps the most that can be said about chemicals that are carcinogenic in laboratory settings is that they may, under some conditions, increase the risk of cancer in humans. Their relative importance in cancer development depends upon many other factors not accounted for in the laboratory experiment. 

---