The theoretical and empirical case for in vitro developmental toxicity screens, and potential applications.

In vitro assays for the screening of developmental toxicity potential have been under development for approximately 15 years. During that period, we have learned that assays consisting of primary cultures of embryonic tissues or cells, intact embryos in culture, or free-living embryos are capable of distinguishing between mammalian developmental toxicants and nondevelopmental toxicants with an accuracy of > or = 80%. Despite this level of performance, there is still considerable reluctance among the scientific community to employ these assays for preliminary screening. In this paper, I review the theoretical basis for the predictiveness of these assays, outline the empirical data indicating their utility in screening toxicants, discuss the major limitations of in vitro assays and how they can be managed, and suggest applications for in vitro pre-screens. The embryo-derived assays should work because they continue to develop in vitro, and the underlying cellular and molecular processes driving this development are the same as those in the mammalian embryo in situ, and therefore, susceptible to the same insults. The assays do work, as specific mechanisms of developmental toxicity have been demonstrated in vitro, and because extensive validation studies have shown them to be highly concordant with traditional in vivo screens. The assays are inherently limited by the fact that they do not include all the levels of complexity of the maternal-embryonic unit; however, these limitations can be minimized by thoughtful assay selection, study design, and interpretation. Potential applications are suggested that complement but do not replace in vivo testing. Pre-screens will make product development more efficient and add to our knowledge about the developmental toxicity of previously untested compounds. In vivo screening would still be conducted on all classes of substances that are currently tested for developmental toxicity; however, fewer chemicals with high likelihood of being developmentally toxic, and therefore not appropriate for further commercial consideration, would be evaluated in these costly screens.