Use of Thresholds in Estimating Cancer Effects

Integrating this information into the risk assessment process would mean that a chemical could be treated as a carcinogen if the concentration were above the cancer threshold level, but could be treated as a non-cancer chemical at lower concentrations. Regulating a chemical in this way would allow for more accurate consideration of toxicology when setting a level that is considered safe to humans. 

In the past, due to lack of knowledge regarding cancer potency at low concentrations, a conservative model has been used to identify slope factors from high dose data in animals. 

Probably the most common of these models is the Linearized Multistage Model (LMS). As discussed in chapter 7, this model assumes the cancer potency of a chemical is linear at low doses, as shown on figure 10.1. A slope factor developed using this model will usually lead to an overestimation of the true cancer potency of a chemical. The slope of the line in our example is shown as the dotted line on figure 10.1. The slope of this curve is then used by the U.S. EPA as the slope factor to set safe concentration levels. For arsenic, this concentration is about 45 parts per billion in drinking water. 

Many epidemiological studies (see chapter 5) have attempted to relate exposure to arsenic through drinking water to skin cancer. Studies conducted in Taiwan and 

Based on this information, the dose level at which cancers first start to be seen from arsenic exposure is near the level at which the detoxification pathway is overwhelmed. Only epidemiological studies with arsenic concentrations in drinking water at or above 0.6 mg/L have reported incidence of skin cancers that increase with dose. The lowest estimated dose corresponding with positive studies of 0.011 mg/kg-day is slightly above the metabolic saturation threshold dose. This is solid evidence that arsenic does not act as a carcinogen in drinking water below this level. 

---