Estimated human potency, carcinogens

In order to express the carcinogenic response or potency, a dose descriptor is used, for example the Tumorigenic Dose (TD). The TD is often set at a defined incidence, for example 5%, the TD5, defined as the dose (or concentration) associated with a 5% incidence of mmors. The dose descriptor can serve as the basis for development of an Exposure/Potency Index (EPl), which is the estimated daily human exposure divided by the TD. A calculated EPl of 10 for the TD5 represents a one million-fold difference between the human exposure and that at the lower end of the dose-response curve, on which the estimate of potency is based. 

The report also showed that a statistically significant increase in the incidenee of lung tumors was observed in rats exposed to antimony trioxide by inhalation. It was also noted that there is an increased incidence of lung cancer in rats exposed by inhalation to antimony trisulfide. This observation, coupled with indications that occupational exposure to antimony processing is associated with lung cancers in humans, is qualitative evidence for the carcinogenicity of antimony by inhalation. However, an earlier U.S. EPA analysis concluded that the animal data were insufficient for quantitative estimation of the carcinogenic potency of antimony. 

The threshold of toxicological concern (TTC) concept has been developed to provide criteria for risk assessment decision-making in the absence of detailed information on chemicals. The approach involves estimating a tolerable human exposure value for all chemicals below which there is a very low probability of an appreciable risk to human health (Kroes et al. 2004, 2005), based on their chemical structures, compared to an extensive toxicity database. As utilized by U.S. FDA in their Threshold of Regulation procedure, structural alerts for high-potency carcinogenicity are included, to increase the assurance of safety. 

The EPA s potency estimate for methylene chloride, shown in Table 8.5, is based on animal carcinogenicity data, and is 0.0075 in units of lifetime risk per dose of one mg/(kg b.w. day). (Human data regarding the carcinogenicity of methylene chloride are inconclusive.) One source of methylene chloride exposure is drinking water, where it can come to be present because of its wide use and escape into the general environment. Suppose the average methylene chloride concentration of a particular drinking water supply is 0.050 mg/1. If people drink 2 liters of this water each day, a total of 0.10 mg of methylene... 

In the absence of human or animal oral dose-response data, the relative potency approaches developed by Watson et al. (1989) and U.S. EPA (1991) are considered to be appropriate methods for estimating the tnmorigenic potency of sulfur mustard by the oral route of exposure. The oral slope factor derived by Watson et al. is approximately one order of magnitude less than the one derived from the relative potency estimated by U.S. EPA (1991). In the emerging area of relative potency analysis, a factor of 10 difference represents a good fit. There is no significant difference between the estimates of sulfur mustard carcinogenic potency relative to B(a)P pnblished by Watson et al. (1989) and U.S. EPA (1991). 

Potency estimates derived from such animal studies help to characterize the dose-response relationship at the low-exposure levels to which humans are likely to be exposed and to predict the quantitative estimate of the risks that humans are likely to encounter at ambient exposures. Experimental evidence for various shapes of the dose-response curve for carcinogens showed that reliable high-dose data from human studies contain examples of superlinearity, linearity, and sublinearity. These are also seen in animal studies. But there are no data to indicate the shape of the dose-response relationship corresponding to lifetime risk of one in a million, the insignificant risk level generally used by the regulatory agencies. 

Although activation of the AHR by DLCs is a key event, mechanistic data indicate that AHR-mediated responses are not well conserved across species, with lower sensitivity in humans. A TEF value for a DLC based on rodent data may overestimate the potency of a DLC in humans, and this has not been considered in the current risk assessment of DLCs. Thus, the current TEF-Toxic Equivalency Quotient scheme tends to compound the conservative estimates of risk that exist within standard risk assessment approaches. Moreover mechanistic differences will now be considered by US EPA in the risk assessment of chemical carcinogens. The mechanistic data currently available for receptor-mediated DLCs and PPs clearly indicate that humans respond differently to these two classes of rodent carcinogens, and these data will need to be incorporated into cancer risk assessments for these chemicals. Full appreciation of the species differences in these receptor mechanisms will require continued development and refinement of models such as primary... 

Rather than regulating carcinogens in products based on estimated cancer risk levels, the carcinogen or product has more often simply been banned in other cases a practical level of control has been estabhshed, with or without promulgation of regulations. The value of the product in commerce appears to have been the major consideration in many cases in others, offsetting public health benefits are more relevant. Carcinogenic potency, its applicabihty to humans, and the amount of exposure may also have been considered, but perhaps in a more qualitative fashion. 

These results challenge some current default approaches to interspecies extrapolation used for risk assessment. In particular, the selection of index animal experiment(s) on which to base the estimation of carcinogenic potency is inconsistent with the observation that there are lognormal distributions of CDio for each chemical in each animal species in fact, no one experiment can be singled out as representative in snch circumstances. Furthermore, the use of allometric scaling for extrapolating quantitative carcinogenicity measures from animals to humans is not supported by the observations in animals, nor the Umited informalion on humans. 

In order to illustrate the various possible approaches to estimating a combined potency for a multisite carcinogen, an example is offered here. The carcinogen selected, 1,3-butadiene, is an important industrial chemical and a widespread air pollutant. Risk assessments for this chemical have been published (OEHHA 1992 US EPA 2002). The more recent estimate by US EPA (2002) depends primarily on estimates of risk from the various human epidemiology studies of industrial exposures to this chemical, but it is their analysis of the animal bioassay data and also that by OEHHA (1992) which are of interest for the present purpose. 

---