Table 3 Criteria for the assay-specific evaluation of the validated in vivo genotoxicity assays. Obligatory criteria are presented in italics. The non-obligatory criteria were applied to evaluate the significance of the acceptable test results

All in vivo assays

 1. Positive and negative control data must be available

 2. An adequate dose range should include at least three doses covering a range from the maximum tolerated dose, 2000 mg/kg body weight for < 14-day or 1000 mg/kg body weight for > 14-day exposures in oral exposure, or lung overload limit (threshold level of particles reached within the lung above which the observed adverse effects may be attributable to particle accumulation and may not reflect a real toxic response) for inhalation exposure, to a dose producing little or no toxicity. The maximum dose may also be defined by toxicity in the target tissue or by a particle concentration which, in a real-life exposure scenario, becomes effectively non-nano due to particle agglomeration

 3. The exposure route should be justified by human exposure

 4. Tissue distribution data should be provided. It should be demonstrated that the material itself, its metabolites or secondary effectors reach the target tissues

Mammalian erythrocyte micronucleus (MN) test

 5. The proportion of reticulocytes (immature erythrocytes, RET%) among total erythrocytes must be determined for each animal by counting a total of at least 500 erythrocytes for bone marrow and 2000 erythrocytes for peripheral blood

 6. At least 4000 immature erythrocytes per animal should be scored for the incidence of micronucleated immature erythrocytes and a minimum of 5 animals per group should be analyzed to reach sufficient statistical power

 7. When peripheral blood is used, it must be established that splenic removal of micronucleated cells from the circulation does not compromise the detection of induced NM in the species selected (this has been clearly demonstrated for mouse and rat peripheral blood)

 8. Sample collection times at which the treatment-related induction of micronucleated immature erythrocytes can be detected are required. In the case of peripheral blood sampling, enough time must also have elapsed for these events to appear in circulating blood. The time required for the nanomaterial to reach the target tissue should be considered in the experimental design. Treatment schedule should be scientifically justified, e.g., with biodistribution data, or by minimum, following the OECD TG. Tests with positive results should be accepted regardless of this criterion

Mammalian bone marrow chromosomal aberration (CA) test

 9. The mitotic index must be determined as a measure of cytotoxicity in at least 1000 cells per animal for all treated animals

 10. At least 200 metaphases per animal (or more if negative control frequency is < 1%) and a minimum of 5 animals per group must be analyzed to reach sufficient statistical power

 11. Treatment schedule should be scientifically justified, e.g., with the help of biodistribution data, or by minimum following the OECD TG. For rodents, the first sampling interval should be the time necessary to complete 1.5 normal cell cycle lengths after the exposure of the target tissue. The time required for the nanomaterial to reach the target tissue as well as its effect on cell cycle kinetics can affect the optimum time for CA detection and should be considered in the experimental design. A later sample collection 24 h after the first sampling time is recommended for the highest dose. Tests with positive results should be accepted regardless of this criterion

Mammalian alkaline comet assay

 12. Automated or semi-automated (quantitative) scoring required, % tail DNA from at least 150 cells per sample (excluding hedgehogs) and a minimum of 5 animals per group are required

 13. For positive results, an examination of one or more indicators of cytotoxicity (e.g. histopathology or trypan blue exclusion) is required. Increases in DNA migration in the presence of clear evidence of cytotoxicity should be interpreted with caution

 14. According to the OECD TG, the samples should be collected no later than 6 h after the final treatment. However, the time required for the nanomaterial to reach the target tissue and biopersistence of the materials should be considered, and positive results accepted with any schedule. For negative results the sample collection schedule should be justified preferably by biodistribution data, and it should be considered that the OECD TG recommends two or more treatments at approximately 24 h intervals

Erythrocyte Pig-a gene mutation assay

 15. A sufficient number of cells to detect at least one mutant cell per sample and a minimum of six animals per group must be analyzed

 16. For optimal sensitivity, a 28-day repeated exposure is recommended, and samples should be analyzed at least once within a few days after cessation of exposure (days 29–31 after the first dose). Other treatment schedules should be scientifically justified. Tests with positive results should be accepted regardless of this criterion