How valid are "old" repeated dose toxicity studies for chemical risk assessment?

An essential endpoint for assessing human health effects also within the REACH process is the repeated toxicity in rodents. In many cases experimental studies are already available. However, these studies often have not been performed according to (current) guidelines, or without GLP. In the EU funded project OSIRIS it was the question, if the NOAELs and LOAELs derived from these old studies are valid, or if the studies must be repeated. To answer this question the RepDose database on repeated dose toxicity in rodents (fraunhofer-repdose.de, Bitsch et al., 2006) created by Fraunhofer ITEM with funding from Cefic LRIwas analysed. As expected the number of target organs increases with the study quality. The most frequently affected targets are liver, kidney, haematology, clinical chemistry and clinical symptoms or body weight is reduced. In inhalation studies nose and lung are also affected quite often. Organs frequently missing in older studies are e.g. testes, uterus and thymus. Regarding the validity of the LOAEL of old studies, it turned out that in >60% of the guideline studies already at the LOAEL several organs are affected. If only one target organ triggers the LOAEL in >80% of the cases it is one of the frequently affected target organs. Based on these findings it can be concluded that in most cases also older studies can be used to derive a reliable NOAEL or LOAEL, if the frequently affected target organs of the respective guideline study have been investigated. A repetition of studies should be considered carefully.