Tilson H A, MacPhail R C, Moser V C, Becking G C, Cuomo V, Frantík E, Kulig B M, Winneke G
Neurotoxicology Division, US Environmental Protection Agency, Research Triangle Park, NC 27711, USA.
Neurotoxicology. 1997;18(4):1065-9.
In the International Programme on Chemical Safety (IPCS) Collaborative Study on Neurobehavioral Screening Methods, eight participating laboratories used a standard battery of behavioral tests to determine, in rats, the effects of seven representative chemicals following acute and repeated dosing. The results of the collaborative study indicate good agreement across laboratories with regard to the data collected in vehicle controls. It was clear, however, that some behavioral measures had significantly more variability than other tests. The laboratories also demonstrated the ability to detect known neurotoxic chemicals and identify profiles of effects that differed from non-neurotoxic agents. The results of the study suggest that appropriate training of personnel is crucial to ensure the reliability of the test battery. The results also underscore the importance of dose selection in behavioral screening studies, since it is sometimes difficult to determine the specificity of behavioral changes in animals receiving high doses of some chemicals. The collaborative study also emphasizes the need to utilize a battery of tests in screening a wide range of potential neurotoxic agents. Analysis of data from such studies poses unique challenges due to the large number of tests and test times, and the consequent possibility of false positives. Some statistical concerns may be alleviated by grouping the results from tests that measure similar functions into neurobiological domains. Although this approach improves confidence in the biological relevance of chemical-induced changes in behavior, it may also lead to false negatives. The exploration of other statistical approaches to analyze data from experiments using a test battery is encouraged. Nevertheless, results of the collaborative study strongly support the use of behavioral tests in hazard identification.
在国际化学品安全计划（IPCS）关于神经行为筛查方法的协作研究中，八个参与实验室使用一套标准的行为测试，来确定七种代表性化学物质在急性和重复给药后对大鼠的影响。协作研究结果表明，各实验室在收集的赋形剂对照数据方面具有良好的一致性。然而，很明显，一些行为测量的变异性明显高于其他测试。各实验室还展示了检测已知神经毒性化学物质以及识别与非神经毒性物质不同的效应特征的能力。研究结果表明，对人员进行适当培训对于确保测试组的可靠性至关重要。结果还强调了行为筛查研究中剂量选择的重要性，因为有时很难确定接受高剂量某些化学物质的动物行为变化的特异性。协作研究还强调了在筛查多种潜在神经毒性物质时使用一组测试的必要性。由于测试数量众多、测试时间长，以及随之而来的假阳性可能性，对此类研究数据的分析带来了独特的挑战。通过将测量相似功能的测试结果分组到神经生物学领域，一些统计问题可能会得到缓解。尽管这种方法提高了对化学物质诱导行为变化的生物学相关性的信心，但也可能导致假阴性。鼓励探索其他统计方法来分析使用测试组进行的实验数据。尽管如此，协作研究结果有力地支持了在危害识别中使用行为测试。
