Report of the BTS/UKEMS Working Group on Dose Setting in In-Vivo Mutagenicity Assays

The BTS/UKEMS Working Group was established to consider dose selection for in-vivo mutagenicity assays, and the feasibility of establishing criteria for identifying maximum dose levels that did not involve relating these to a high fraction (50-80%) of the estimated LD50 value.

In view of the importance attached by regulatory authorities to negative results from in-vivo tests, namely reassurance that mutagenic potential seen in vitro could not be expressed in the whole animal, the need for the use of some form of MTD was accepted. The crucial question facing the group was whether the use of 'evident toxicity' rather than some index of lethality would result in any meaningful loss of sensitivity of the assays.

The group endorsed the concept of a limit dose for relatively non-toxic compounds and supported the use of a value of 2 g kg-1 for single oral dosing and 1 g kg-1 for repeated dosing, in line with the general values used by the OECD and EEC.

In order to assess the question of sensitivity of the assays the group considered the available data from the published literature, and from 'in-house' studies, on dose-response for mutagenicity and for toxicity, using the same dosing regime. It rapidly became apparent that few data were available, and that these were limited essentially to the micronucleus test; there were inadequate data to consider any other methods. In addition, there was the added complication that most of the available data related to protocols which were less rigorous than those currently recommended. The group thus concentrated on the micronucleus test because of its relatively wide use and since it had given rise to specific concerns due to a recent recommendation from the relevant EPA Gen-Tox group namely that the top dose should be 50-80% of the estimated LD50 value. It was noted that the EPA appear to be considering this approach as one alternative when dose setting, with the possibility of the use of a dose producing overt toxicity as another. Available data indicate that around 90% of tested mutagens would have been identified using an MTD based on non-lethal criteria. Moreover the percentage would be expected to be higher if all tests had been carried out to current protocol standards. However, the possibility of missing compounds could not be completely eliminated.

Furthermore, it was important to put the bone marrow study in context. In the UK, Regulatory Authorities would not accept negative data from one tissue as providing adequate reassurance regarding the absence of in-vivo activity. Data from at least one other assay using a different tissue would be needed. It would not therefore be necessary to use 'heroic' and unrealistically high doses in the bone marrow assay in a misguided attempt to obtain absolute assurance from the one study. It is believed that Regulatory Authorities in most other countries would seek data from more than one in-vivo assay before discounting positive data from in-vitro studies.

The group also considered in quantitative terms, the actual difference in MTD in the mouse if based on 'evident toxicity' or on lethality (an estimate of a dose equivalent to 50-80% the LD50 value). There was relatively little difference between the two levels, due to the steep dose response for toxicity seen in the mouse with most compounds.