Mutagenicity studies are highly specialized. There are multiple hereditary components in both somatic and germinal cells that may be affected by drugs. During the 1970s, it was thought (somewhat naively) that these studies may be replacements for the long and costly carcinogenicity studies that are required for many drugs. Although this goal was never realized, mutagenicity studies nonetheless provide useful indications of the ability of a drug to alter genetic material, which may later be manifested in studies of carcinogenic or terato-genic effects (Kowalski, 2001). Genotoxicity studies are relatively inexpensive and may also serve, early in the drug development process, to assure drug developers and regulators that no obvious risk of such adverse effects exists, albeit knowing that more definitive studies to evaluate terato-genic and carcinogenic effects will not come until later.
An exhaustive review of the various components of a mutagenicity evaluation will not be attempted here. Multiple guidelines are available. Those issued by the ICH include general guidelines (Federal Register April 24 1996) and specifics related to the core battery of studies required (Federal Register April 3 1997). Tennant et al (1986) have summarized the correlation between the results of a battery of mutagenicity assays and the probability of the material producing a positive carcinogenic response in long-term rodent studies. Obviously, mutagenicity studies cannot address issues of non-genetic carcinogenicity or teratogeni-city.
Positive results in one or more mutagenicity assays do not necessarily translate into human risks. Mechanistic studies may show that such responses would not occur in the human cell population, or the concentrations at which positive responses occurred may far exceed any concentration of drug that may occur in the clinical setting. Many drugs are on the market today that have produced some type of positive response in these studies and yet it has been concluded that no human risk is present or the potential risk is not known (e.g. aspirin causes chromosomal breaks). A Fairly standard worked example is provided by Fox et al, 1996.
Was this article helpful?