EPA and other competent regulatory agencies around the world recognize that advances in biology and computer science provide opportunities to make the risk assessment process more informative and efficient than is currently possible with traditional approaches, largely based on animal bioassays. These have been described in a variety of notices following the NAS's 2007 report "Toxicity Testing in the 21st Century, A Vision and a strategy" and are beginning to be formalized (e.g., OECD AOP program https://www.oecd.org/chemicalsafety/testing/adverse-outcome-pathways-molecular-screening-and-toxicogenomics.htm). How, and to what extent, do you recommend that such approaches be applied to further evaluate carcinogenic potential and increase confidence in the conclusion that 1,3-D is not likely to be carcinogenic to humans?

The OECD AOP program may result in better information on the mode of action or mechanisms of toxicity for various compounds. For example, if we had a better understanding of the mechanism by which 1,3-D increased the incidence of benign lung tumors in B6C3F1 male mice, we may be better able to predict the probability that this compound could cause lung tumors in humans. The problem is that we are not there yet. What we have are empirical observations in animals that we have to attempt to extrapolate to humans. Without the benefit of mechanistic information on the observed effects of 1,3-D in mice and rats, we must still decide 1) are the observed effects real in these species and 2) to what extent can we extrapolate these effects to humans? Based on where we are today, the weight of evidence suggests that 1,3-D is unlikely to be carcinogenic to humans.

I think that the AOP approach proposed by OECD is interesting. I think that in this particular case of 1,3-D the AOP approach would organize results and give further information in addition to all the results obtained by in vitro and in vivo studies, MOSTLY if we consider the molecular initiating event (MIE) and the outcomes.

The absence of human relevance of the lung tumors could be better supported when an AOP for the response can be developed (may have been done with styrene ?) and individual steps can not be demonstrated to occur with 1,3-D. This however, may require a significant research effort that may be difficult to justify if the tumors are not treatment related .