Reduction of animal use for toxicological evaluation: the Netherlands Health Council report on the LD50

In 1981 the Health Council of the Netherlands installed a committee to report on the scientific value of the LD50 and on ways of obtaining the information needed for the toxicological evaluation of chemical products at the expense of less experimental animals and of less suffering. This was done at the request of the Minister of Public Health and Environmental Hygiene. The first four reports have been published. The first one discusses the different methods for estimating the median lethal dose, and stresses that a high precision is seldom needed. A limit test would suffice in many cases. The test for abnormal toxicity of the European Pharmacopoeia (General method) has a power of discrimination of 0.96 when the dose injected in fact is equal to the LD50; the power of the version of this test intended for immunosera and vaccines is even higher. The second report deals with pharmaceutical preparations, immunosera and vaccines. The types of toxicity data and the precision required differ in the successive stages of screening, selection, development up to registration and quality control. Determination of an LD50 with a high precision is required only for preparations with a narrow therapeutic margin, and in very exceptional cases for quality control.