For many medical devices, ISO 10993-1:2018 and FDA Guidance (Use of International Standard ISO 10993-1, “Biological Evaluation of medical devices – Part 1: Evaluation and testing within a risk management process,”) both recommend genotoxicity as an endpoint for evaluation. But what is the purpose of these assays and why are they necessary?
Purpose of Genotoxicity Assays
ISO 10993-3, Biological evaluation of medical devices – Part 3: Tests for genotoxicity, carcinogenicity and reproductive toxicity, outlines the general requirements for genotoxicity evaluation of medical devices. According to the standard, the purpose of genotoxicity testing is “to detect the two major classes of genetic damage:
- Gene mutations (point mutations);
- Chromosomal damage [structural aberrations such as translocations, small or large deletions and insertions, and numerical chromosomal aberrations (aneuploidy)].”
The standard goes on to specify that in vitro “bacterial reverse mutation assays have been shown to detect relevant genetic changes produced by the majority of genotoxic carcinogens detected by rodent assays;” and as such, it is recommended that all test batteries for genotoxicity testing include a bacterial reverse mutation assay.
What is the Bacterial Reverse Mutation Assay?
The bacterial reverse mutation assay, commonly referred to as the Ames Assay, was developed by Bruce Ames in the early 1970s and was designed to aid in the detection of carcinogenic and mutagenic compounds. To detect point mutations that could lead to genotoxicity or carcinogenicity, the assay exposes specific strains of Salmonella typhimurium and Escherichia coli to medical device extracts in both the presence and absence of metabolic activation. Therefore, if the device extract were to contain enough of an extractable compound to cause DNA base pair substitution, addition, or deletion, the assay would yield a positive result.
For medical devices, guidelines describing how to perform the Ames assay in detail are provided in ISO/TR 10993-33 and OECD 471. However, in reviewing many Ames assay reports, there are several key areas in which we most often identify significant gaps in the testing methods. These include the bacterial strains selected, the positive controls selected for use in the absence of metabolic activation, and the positive control(s) selected for use in the presence of metabolic activation.
As described in OECD 471 the test requires the use of at least five strains of bacteria. It is recommended to use the following combination of strains:
- S. typhimurium TA1535
- S. typhimurium TA1537 or TA97 or TA97a
- S. typhimurium TA98
- S. typhimurium TA100
- E. coli WP2 uvrA, or E. coli WP2 uvrA (pkM101), or S. typhimurium TA102
Each strain has slightly different properties and, as such, are able to aid in the detection of different types of mutagenic compounds. For example, S. typhimurium TA1535, TA1537, TA97a, TA97, TA98, and TA100 may not be able to detect some oxidizing mutagens, cross-linking agents, or hydrazines. This is in part due to these strains containing GC base pairs at the primary reversion site. However, as E. coli WP2 strains and S. typhimurium TA102 have an AT base pair at that site, including one of these strains in the assay allows for the detection of mutagens otherwise not observed.
Another important consideration in the Ames assay is the use of positive controls. Given there are multiple bacterial strains and each is able to detect slightly different genotoxic compounds, there is no “one size fits all” approach. Therefore, it is important to carefully select the positive controls based on the specific strains being used in the assay. Examples of strain-specific positive controls to be used in the absence of metabolic activation are listed in OECD 471 and are reproduced in Table 1 below.
Table 1. Strain-specific Positive Controls to be Used Without Metabolic Activation
| Chemical | CAS No. | Compatible Bacterial Strain |
| Sodium azide | 26628-22-8 | TA1535 and TA100 |
| 2-Nitrofluorene | 607-57-8 | TA98 |
| 9-Aminoacridine | 90-45-9 | TA1537, TA97, and TA97a |
| ICR191 | 17070-45-0 | |
| Cumene hydroperoxide | 80-15-9 | TA102 |
| Mitomycin C | 50-07-7 | WP2 uvrA and TA102 |
| N-Ethyl-N-nitro-N-nitrosoguanidine | 4245-77-6 | WP2, WP2 uvrA, and WP2 uvrA (pKM101) |
| N-Methyl-N-nitro-N-nitrosoguanidine | 70-25-7 | |
| 4-Nitroquinoline 1-oxide | 56-57-5 | |
| Furylfuramide (AF-2) | 3688-53-7 | Plasmid-containing strains |
As stated above, the Ames assay is to be performed both with and without metabolic activation. Due to humans and bacteria metabolizing some compounds differently, the use of adding a mixture, such as S9, to induce metabolic activation is necessary, given some compounds require metabolization to induce genotoxicity. As such, it is necessary to include at least one positive control to ensure the efficacy of the S9-mix. Examples of acceptable positive controls for metabolic activation include the following:
- 9,10-Dimethylanthracene (CAS No. 781-43-1)
- 7,12-Dimethylbenzanthracene (CAS No. 57-97-6)
- Congo Red (CAS No. 573-58-0; for the reductive metabolic activation method)
- Benzo(a)pyrene (CAS No. 50-32-8)
- Cyclophophamide (monohydrate) (CAS No. 50-18-0; CAS No. 6055-19-2)
- 2-Aminoanthracene (CAS No. 613-13-8)
However, one common gap observed is the use of 2-aminoanthracene as the sole positive control for metabolic activation. As described in OECD 471, the use of 2-aminoanthracene alone is not sufficient and, if it is used, an additional chemical should be used, such benzo(a)pyrene or dimethylbenzanthracene.
Frequently Asked Questions (FAQ)
Do I need to perform the Ames Assay to evaluate the genotoxicity of my device?
First, it is important to determine if genotoxicity is a necessary endpoint for evaluation for your device. This can be done by referencing Table A.1 in ISO 10993-1:2018 and Table A.1 of the 2023 FDA Biocompatibility Guidance document, if you intend to submit in the United States. If the endpoint is not required for evaluation, based on the device categorization and there are no known genotoxic compounds present in the device, then additional evaluation of genotoxicity is not necessary. If the device categorization indicates genotoxicity is a recommended endpoint for evaluation, then genotoxicity testing may need to be considered. However, chemical characterization testing performed in alignment with ISO 10993-18:2020 and a subsequent toxicological risk assessment (TRA) performed in alignment with ISO 10993-17:2023 and ISO/TS 21726:2019 can often be used to assess the risk of genotoxicity instead. If chemical characterization testing is not performed or if the results of the chemical characterization testing and TRA do not support a negligible risk of genotoxicity, then additional genotoxicity testing may be necessary to perform.
Can I perform chemical characterization testing and the Ames assay?
According to ISO 10993-1:2018, it is generally recommended to perform an evaluation of the materials of construction and any manufacturing processes or processing aids prior to any chemical or biological testing. Following this evaluation, it is recommended to first perform any necessary chemistry testing before any biological testing. The main purpose for this process is to have a clear understanding of any risks before performing testing in any animals or using the device in humans. However, given the Ames assay is performed in bacterial cells, this assay is not performed on animals. As such, yes, the assay can be performed simultaneously with the chemical characterization testing without conflicting with ISO 10993-2:2022.
If the results of the Ames assay for my device are positive, does this mean I need to perform a 2-year carcinogenicity assay to determine if there is any risk?
Not necessarily. Prior to recommending carcinogenicity testing, it will be important to assess which bacterial strains had positive results and if those positive results were in the presence or absence of metabolic activation. This information will help narrow down the type of genotoxic mechanism of the potential extractable from the device. It will also be important to evaluate the other biological testing results, as well as the known materials of construction and any manufacturing processes or processing aids. Did previous cytotoxicity testing fail and could the observation be a cytotoxic response instead of a genotoxic response? Could there be a residual from the manufacturing processes that is causing the positive result? If so, maybe implementing an additional cleaning step and reperforming the Ames assay would be sufficient to mitigate the risk.
If it does not appear that this could have been a false positive result and it is unknown where the genotoxic substance could be originating, chemical characterization testing and a TRA might be recommended prior to performing carcinogenicity testing (assuming it has not already been performed). The detected extractables would then be evaluated to determine if any present a genotoxic risk, specifically via the same mechanisms of action indicated by the positive Ames assay results. If no genotoxic compounds are detected, then other avenues may be evaluated to determine the cause of the positive result. If some genotoxic/carcinogenic compounds are detected, an additional assessment of carcinogenicity risk may need to be performed in the TRA to address the level risk to a patient.
