Main gaps identified: high rate of misleading/false genotoxic positives
Existing in vitro methods, whilst having a high sensitivity (and thus low false negative rate), have a relatively low specificity and thus high rate of false (“misleading”) positive results, which typically has led to unnecessary follow-up testing in vivo for the confirmation of these results.
To address this issue, the recommendations of an ECVAM workshop (Kirkland et al., 2007) have contributed to several international initiatives.
EURL ECVAM strategy to avoid and reduce animal use in genotoxicity testing
Based on an analysis of regulatory requirements for this endpoint within different pieces of EU legislation, EURL ECVAM has proposed a pragmatic approach to improve the traditional genotoxicity testing paradigm that offers solutions in both the short- and medium-term and that draws on the considerable experience of 40 years of regulatory toxicology testing in this area (EURL ECVAM strategy to avoid and reduce animal use in genotoxicity testing, released in December 2013).
In its strategy, EURL ECVAM has considered that efforts should be directed towards the overall improvement of the current testing strategy for better hazard and risk assessment approaches, which either avoids or minimises the use of animals, whilst satisfying regulatory information requirements, irrespective of regulatory context. Several opportunities for the improvement of genotoxicity testing have been identified which aim to:
- enhance the performance of the in vitro testing battery so that fewer in vivo follow-up tests are necessary and
- guide more intelligent in vivo follow-up testing to reduce unnecessary use of animals.
For a review of the achievements and implementation of the EURL ECVAM strategy plan read Corvi and Madia, 2017.
Are there categories of positive results from the Ames test that are irrelevant or signify low risk?
Differently from in vitro mammalian cell tests, no systematic analysis on Ames test accuracy was conducted until recently, despite the Ames test being the primary test for genotoxicity screening and the most used test within the in vitro battery.
Therefore, a workshop was organized by EURL ECVAM to investigate whether the in vitro mammalian cell genotoxicity test results could complement and mitigate the implications of a positive Ames test response for the prediction of in vivo genotoxicity and carcinogenicity, and if patterns of results could be identified (see workshop report Kirkland et al., 2014a).
The workshop recommended to construct a Genotoxicity & Carcinogenicity Consolidated database (DB) which was consequently launched at the end of 2014. The data collected were analysed to investigate whether negative results in mammalian cell tests were associated with absence of carcinogenic or in vivo genotoxic activity despite a positive Ames test (Kirkland et al., 2014b).
Recommended list of chemicals to assess the performance of new or improved genotoxicity tests
Reference chemical selection is a key step in the development, optimisation and validation of alternative test methods. A revised reference list of genotoxic and non-genotoxic chemicals recommended for assessing the performance of new or improved in vitro genotoxicity test methods has been developed to fit the following different sets of characteristics (Kirkland et al., 2016):
Group 1: Chemicals that should be detected as positive in in vitro mammalian cell genotoxicity tests. Chemicals in this group are all in vivo genotoxins at one or more endpoints, either due to DNA-reactive or non DNA-reactive mechanisms. Many are known carcinogens with a mutagenic mode of action, but a sub-class of probable aneugens has been introduced.
Group 2: Chemicals that should give negative results in in vitro mammalian cell genotoxicity tests. Chemicals in this group are usually negative in vivo and non-DNA-reactive. They are either non-carcinogenic or rodent carcinogens with a non-mutagenic mode of action.
Group 3: Chemicals that should give negative results in in vitro mammalian cell genotoxicity tests, but have been reported to induce gene mutations in mouse lymphoma cells, chromosomal aberrations or micronuclei, often at high concentrations or at high levels of cytotoxicity. Chemicals in this group are generally negative in vivo and negative in the Ames test. They are either non-carcinogenic or rodent carcinogens with an accepted non-mutagenic mode of action.
Since substantial in vivo testing is still required by chemical authorities for confirmation of the genotoxic prediction in vitro, it is crucial to address issues related to the reduction and refinement of genotoxicity tests. Recommendations on opportunities to reduce the number of animals in genotoxicity tests are published by Pfuhler and colleagues in an ECVAM workshop report (Pfuhler et al., 2009). Genotoxicity OECD Test Guidelines have recently been revised to also include some animal welfare considerations (OECD TG website).