Accuracy, toxicity tests

Toxicologists must rely on results obtained from an alternative method if it is to serve as a replacement for an in vivo toxicity test. Two measures of alternative method performance must be known in order to define reliability from a test user s point of view. First, a toxicologist must know it is possible to consistently reproduce the data obtained from the alternative method over long periods of time. A test that does not provide the same results on the same test substance repeatedly would not be useful in the safety assessment process. Second, it must be possible to consistently predict in vivo toxicity endpoints at a known level of accuracy and precision. These measures of reliability are objective endpoints that can be measured experimentally. The part of the validation process that provides the data needed to confirm the reliability of an alternative method as proposed by its developers is the validation study. 

A toxicity test is carried out following standard protocols. The accuracy of the test is determined by its sensitivity and specificity. The test s sensitivity determines how many should-be positives are scored as positive, whereas the specificity tells how many inactive substances also score negative in the test. In the example below we suppose that the test has 95% sensitivity and 95% specificity. Suppose that the truth is that 2% of 1000 substances are carcinogenic. However, in spite of the good accuracy of the test, it may be too coarse for screening. Let us do some calculation to illustrate this ... 

Accuracy criteria are not applicable to toxicity testing endpoints, because there are no standard organism responses against which to compare test results. In place of... 

In the hierarchy of toxicity testing, the use of structure-activity relationships is becoming an increasingly important predictive tool. The reliability of this tool is, of course, dependent on the accuracy of the data bases which it incorporates. Efforts are under way to make these bases more reliable. Further steps in safety evaluation may follow a "decision tree" approach any scheme adopted should involve a variety of screening procedures, including tests for genetic toxicity, as well as early studies of metabolism and pharmacokinetics. 

The problem of toxic subjects detection in the tested objects can be solved by two options chemical analysis, for revealing separate toxics, or their products, and biotesting with the result of the tested samples toxicity degree indication without identification of the agent. Qualitative and quantitative chrmical/analytical methods allow with the higher accuracy and, in some cases, rapidly detect presence of the separate toxics or their products in the tested objects. It is important for the regular detection of the different pollutions of any agents in the tested objects. 

Analytical techniques have gone through considerable changes in the past 20 years. With the development of more sensitive and selective analytical instrumentation the analyst has been able to detect and identify minute quantities of materials never before seen. This has brought about a keen awareness of the widespread distribution of toxic hazards and also the need to study the long term effects of low level exposures. The development of new methodology is a dynamic process. However, new methods should always be thoroughly tested to demonstrate the precision and accuracy of the results obtained. 

Standardization Imposition of rules permitting to check or validate the accuracy of a test using live organisms. For example, the use of a well-defined experimental procedure and the use of a reference toxicant are important rules to standardize a test. Test standardization also requires that the test be feasible by many laboratories and yield comparable results with the same test substance. Volume 1(14). 

Stay FS, Flum TE, Shannon LJ, Yount JD. 1989. An assessment of the precision and accuracy of SAM and MFC microcosms exposed to toxicants. In Cowgill UM, Williams LR, editors. Aquatic toxicology and hazard assessment. ASTM STP 1027, Volume 12. Philadelphia (PA) American Society of Testing and Materials, p 189-203. 

---