Carcinogenicity Tests Animal Bioassays

As the mechanism of carcinogenesis in both humans and animals is not well understood, the only acceptable procedure for determining whether a chemical is likely to be a carcinogen is the examination of experimental animals exposed to the suspect material under carefully controlled conditions. This procedure relies on the assumption that animals will behave in essentially the same way as humans to carcinogen exposure, i.e., the mechanism of tumor induction will be similar in both animals and humans. Mechanistically based, short-term tests for carcinogenicity prediction not involving experimental animals are still a distant and elusive goal.

The basic approach for carcinogenicity testing involves administering the test material to two suitable animal species for a considerable proportion of their natural lifespan. Because of their small size and relatively short life expectancy, the rat and mouse are the species of choice, although the hamster is occasionally used. In the US, inbred strains of animals are widely used (the F344 rat and the B6C3F1 hybrid mouse), although out-bred strains are commonly used in Europe. To examine the carcinogenic potential of food components, the test substance is usually given in the diet, although in some circumstances administration may be in the drinking water or by gavage. The study continues until a certain proportion in one or other of the treatment groups has died or has been killed in a moribund state. As a minimum, 50 animals are allocated at random to each of the experimental groups, allowing a statistically significant carcinogenic effect to be detected if five animals in a test group develop tumors and no animals in the control group do.

During the study, the animal's clinical state is regularly monitored and at the end of the study a complete necropsy is performed on all surviving animals. Any tumors found are classified as either neoplastic or non-neoplastic and some attempt is made to determine whether any tumors seen were the cause of the (early) death of the animal (fatal tumors) or were unrelated to the death (incidental tumors). The procedures of these bioassays are conducted under rigorous conditions defined by the Code of Good Laboratory Practice (GLP).

Tests are essentially of two types: the first, used widely under the National Toxicology Program (NTP) in the US, is designed to examine the ability of the test material to induce cancer in the species used; the second, is aimed at determining the cancer incidence in respect of dose - a classical dose-response study. The former requires a few treatment groups, including a relatively high-dose group in order to maximize the chance of detecting a carcinogenic effect, whereas the latter requires a wide range of dose groups to define accurately the dose-response relationship.

The analysis of a carcinogenicity bioassay is aimed at determining whether the administration of the test chemical has resulted in an increase in the incidence of tumors at one or more sites. In order to accomplish this analysis, two major confounding factors may have to be taken into consideration. The first is the effect of differences in mortality rates between the control and treated groups and the second is the effect of differences in food intake and its consequence on body weight. Both factors can substantially alter the tumor pattern observed in different groups. Early deaths may prevent the animals reaching tumor-bearing age, and reduced food intake and the associated reduction in body weight may result in a considerable reduction in tumor incidence.

The interpretation of the results of a bioassay are complex but most authorities work to the 'weight of evidence' principle. This evidence is taken in the light of the 'adequacy' of the bioas-say, which is dependent on some of the factors previously discussed. Strong evidence for the compound being a genotoxic carcinogen would be increased malignant tumor incidence in two species, with tumors at multiple sites showing a clear dose-response relationship. Rare or unusual tumors at a site would be given added weight. Equivocal evidence may result from a statistically marginal result or only an increase in commonly occurring benign tumors. Tumor development in only one species and in association with species-specific toxicity is characteristic of nongenotoxic (or epigenetic) carcinogens. Sometimes, problems associated with such findings may be clarified by further mechanistic studies or by reference to historical data. When the data from bioassays are considered in human risk assessment, other factors must clearly also be taken into consideration. These may include evidence of genotoxicity in short-term tests and data on metabolism and potential human exposure. Furthermore, a measure of risk at doses substantially below the bioas-say dose may be needed. This may require an extrapolation using mathematical models. As yet no general agreement has been reached as to the most appropriate method, and so the calculated risk given by different methods may vary considerably. Thus, the final assessment may be made on quite pragmatic grounds, in which the experience and expertise of a number of individuals are drawn on to reach a consensus opinion.

Short-Term Predictive Tests

A large number of short-term tests have been developed in an attempt to predict carcinogenic potential and thereby reduce the reliance on animal tests. These include assays for detecting gene mutation, damage to chromosomes, or damage to the whole genome.

Gene mutation can be assessed in bacteria, yeasts, or mammalian cells in culture (see Table 4). Since many of the cell systems used are unable to activate metabolically the majority of test chemicals, an exogenous mammalian metabolizing system, the so-called S-9 mix, is incorporated into the assay. Chromosome damage can be measured in cell lines in vitro or by using animals exposed for a short time to the chemical. Structural damage produced can include chromosome and chromatid gaps and breaks, rings, fragments, dicentrics, translocations, and inversions. A short-term in vivo assay measuring

Table 4 Short-term test systems for predicting carcinogenic potential

Test system

Cell used

End point

Bacterial mutation

Salmonella typhimurium TA strains

Reversion to histidine independence

Escherichia coli WP2

Mammalian gene mutation

Chinese hamster lung (V79)

Loss of HPRT, TK, or Na+/K+ ATPase

Chinese hamster ovary (CHO)


Mouse lymphoma (L5178Y)

Human transformed lymphoblastoid (TK6)

Chromosome aberration in vitro

Chinese hamster fibroblast (CHL)

Chromosome/chromatid aberration

Chinese hamster ovary (CHO)

(gaps, breaks, deletions)

Human peripheral blood lymphocytes(PBL)

Chromosome damage in vivo

Bone marrow erythrocytes (mouse)

Micronuclei induction

Heritable damage in vivo

Rodent germ cells

Dominant/lethal mutations; heritable

translocations, etc.

HPRT, hypoxanthine phosphoribosyl transferase; TK, thymidine kinase.

HPRT, hypoxanthine phosphoribosyl transferase; TK, thymidine kinase.

unscheduled DNA synthesis (UDS) in rat liver or gut is recommended by most regulatory authorities if there is a positive response in any in vitro assay and a negative response in an in vivo cytogenetics assay. Other test methods and end points are under consideration by regulatory authorities as indicators of genotoxic potential including the COMET assay for assessing DNA damage, and aneuploidy, the change in chromosome number resulting from damage to the cellular architecture (spindle) controlling chromosome replication.

The last two decades have seen extensive efforts to determine whether short-term tests are suitable for predicting carcinogenic potential. The early validation studies suggested good predictability, with correct identification of over 90% of carcinogens (high sensitivity) and over 90% of noncarcinogens (high specificity). In later evaluations, a much lower figure (60%) was obtained. However, when carcinogens known to react by nongenotoxic mechanisms (e.g., hormones or peroxisome proliferators) were excluded, the predictability was improved suggesting that short-term tests are suitable for detecting those carcinogens that act by a genotoxic mechanism.

Although many regulatory authorities have guidelines for carcinogenicity evaluation, which include short-term tests, they all still require animal studies as the ultimate test for carcinogenicity. However, the use made of short-term tests varies. In the US, the Food and Drugs Administration (FDA) recommends a battery of short-term tests for all 'additives' for which cumulative dietary intake is expected to exceed 1.5 mg per person per day in order to assist in the interpretation of animal feeding studies. Some expert bodies, such as The International Agency for Research in Cancer, use short-term tests as an adjunct to animal carcino-genicity studies in their evaluation process, giving added weighting in their assessment of likely human hazard to an animal carcinogen that is also positive in short-term tests.

However, until a consensus can be reached as to what a positive or negative result in an animal feeding study means in terms of whether the compound may or may not be a human carcinogen, the further development of better (faster/cheaper) short-term tests may be a futile exercise.

Weight Loss All Star

Weight Loss All Star

Are you looking to lose weight or even just tone up? What is stopping you from having the body you want and help you feel great at the same time? I created Weight Loss All-Star for all those wanting to lose weight, and keep the weight off. I know how hard it is to do diets and stick with them, and get the motivation to get up and exercise.

Get My Free Ebook

Post a comment