Comment submitted by D. Eastmond

I am writing to provide public comments on the new EPA document (EPA 120/R-07/002-A) entitled, ?Framework for Determining a Mutagenic Mode of Action for Carcinogenicity: Using EPA?s 2005 Cancer Guidelines and Supplemental Guidance for Assessing Susceptibility from Early-Life Exposure to Carcinogens?. In general, I think the proposed framework represents a reasonable and pragmatic approach to determine whether mutagenicity plays a significant role in the mode of action of environmental chemicals. However, I do have concerns about certain sections of the framework. My comments on specific sections are outlined below: 1) There is increasing evidence that the relationship between dose and mutagenic response for some mutagenic agents is linear and for some it is non-linear. For agents that exhibit a steep non-linear relationship, the use of linear extrapolation and an additional safety or uncertainty factor to protect against early life exposures may not be warranted. To some degree, this has been acknowledged in the document in recognizing that agents that are uniquely positive for inducing aneuploidy are not recommended to be classified as acting through a mutagenic mode of action. However, I believe that a similar situation will hold true for some DNA-reactive mutagenic chemicals. For examples, see articles by Doak et al. (2007) and Jansen et al., 1995. At this point, we do not know enough to accurately predict which chemicals will fall into each category. However, it would be useful if the framework were written in such a way as to accommodate different classes of mutagenic chemicals (i.e. those that exhibit either linear or non-linear dose response relationships) as they become identified and as the properties that differentiate between classes become convincingly established. Personally, I believe that a health protective approach should be used as the standard or default approach but would hope that the framework would recognize that, with the presentation of convincing evidence, some mutagenic modes of action may not pose the same risks and should be regulated differently. 2) I believe that the approaches outlined in Figure 1 versions 1 and 2 are flawed as they provide little incentive for the regulated party to conduct the studies necessary to determine whether an agent is acting through a mutagenic mode of action. In circumstances in which the risk assessor or manager is unable to determine whether a carcinogenic chemical acts through a mutagenic mode of action, the arrow should lead to a tentative conclusion that the agent acts through a mutagenic mode of action. Otherwise, the manufacturer or importer has no incentive to conduct additional studies and in fact, has a strong disincentive to perform the additional research that would address the issue and answer these questions. 3) I believe that a situation in which a tumor exhibits a high frequency of specific gene mutations or exhibits a unique mutational spectrum that is consistent with the known mutagenic activity of a chemical can be useful in deciding whether a chemical acts through a mutagenic mode of action. I believe that the text at the bottom of Figure 1 v.2 and elsewhere in the text such as pages 24 and 30 should be modified to reflect this. 4) The document is inconsistent in the way that it treats aneuploidy and aneugenic chemicals. The term mutagenic is defined in such as way as to exclude aneugens. Yet, these agents are used as an example of mutagenic agents that may not act through a mutagenic mode of action. I would prefer to have mutagenic agents subdivided into separate subcategories such as 1) directly mutagenic agents (DNA reactive), 2) indirectly mutagenic agents (such as those that act through the generation of reactive oxygen species) and mutagenic agents that act through protein targets (such as aneugens, topoisomerase II inhibitors, etc.). The approach to be used for each of these classes would then be described. 5) While mutagenic agents are frequently active in multiple genotoxicity tests, I believe that a significant, reproducible response in a single in vivo assay should be sufficient to classify an agent as a mutagenic agent and may warrant its classification as acting through a mutagenic mode of action. Positive results in multiple genotoxicity endpoints should not be required. 6) I believe that the report by Thybaud et al (1997) was part of the International Working Group on Genotoxicity Testing and was not formally conducted under the auspices of the International Association of Environmental Mutagen Societies as described on page 21, line 18. 7) Page 22 lines 10-11. The sentence is incomplete and doesn?t make sense as written. 8) Page 24, line 28. Whether or not a tumor response occurs early in a chronic study may be more a reflection of dose than whether an agent acts through a mutagenic mode of action. High doses shorten the latency period whereas the latency period may be much longer for low doses of mutagenic compounds. For an example see the results of the 2-acetylaminofluorene (?megamouse?) study conducted by the NCTR. 9) I believe the definition of mutagenic presented on page 8, line 24-25 is unduly narrow. I would suggest using the following definition of a mutation. A mutation is a heritable alteration in the genetic composition of the cell. This definition encompasses a broader range of mutagenic events including aneuploidy. References: 1: Doak SH, Jenkins GJ, Johnson GE, Quick E, Parry EM, Parry JM. Mechanistic influences for mutation induction curves after exposure to DNA-reactive carcinogens. Cancer Res. 2007 Apr 15;67(8):3904-11. 2: Jansen JG, Vrieling H, van Teijlingen CM, Mohn GR, Tates AD, van Zeeland AA. Marked differences in the role of O6-alkylguanine in hprt mutagenesis in T-lymphocytes of rats exposed in vivo to ethylmethanesulfonate, N-(2-hydroxyethyl)-N-nitrosourea, or N-ethyl-N-nitrosourea. Cancer Res. 1995 May 1;55(9):1875-82.