Nothing is perfect, not even the local lymph node assay: a commentary and the implications for REACH.

For many regulatory authorities, the local lymph node assay (LLNA) is the preferred assay for the predictive identification of skin-sensitizing chemicals. It is the initial requirement for sensitization testing within the new REACH (Registration, Evaluation, Authorization and Restriction of Chemical substances) regulations in the European Union. The primary reasons for the preferment of the LLNA are the animal welfare benefits it provides compared with traditional guinea-pig methods (refinement and reduction of animal usage) and the general performance characteristics of the assay with regard to overall reliability, accuracy, and interpretation. Moreover, a substantial published literature on the LLNA is available making it appropriate for use as a benchmark against which new approaches, including in vitro alternatives, can be evaluated and validated. There is, therefore, a view that the LLNA represents the 'gold standard' for skin sensitization testing. However, although this is probably correct, it is important to recognize and acknowledge that in common with all other predictive tests (whether they be validated or not), the LLNA has limitations, in addition to strengths, some of which were mentioned above. Arguably, it is the limitations (e.g., the occurrence of false positive and false negative results) of test methods that are most important to understand. With respect to the LLNA, these limitations are similar to those associated with guinea-pig skin sensitization methods. Among these are the occurrence of false positive and false negative results, susceptibility of results to changes in vehicle, and the possibility that interspecies differences may confound interpretation. In this commentary, these issues are reviewed and their impact on the utility of the LLNA for identification, classification, and potency assessment of skin sensitizers are considered. In addition, their relevance for the future development and validation of novel in vitro and in silico alternatives is explored.