3. Cancer Risk: The contribution of chemicals to the cancer burden

The presence of a chemical in our body is not necessarily harmful9 - risk depends on the amount of exposure and the inherent toxicity of the chemical. These factors are the starting points for determining the dose-response relationship – how much exposure to a chemical is needed to increase our risk of disease by 10 percent, 50 percent or more. Estimating these risks can be extremely difficult when the only information available is the concentration of a chemical measured by biomonitoring.

One potential solution is the use of biomonitoring equivalents, or BEs – – tools that translate measured chemical concentrations to health risks, usually based on government safety standards.10 For cancer, a BE estimates the measured concentration of a chemical that would on average result in a one-in-a-million to 1-in-10,000 chance of getting cancer, the usual standards used in EPA risk assessments.

How much risk do carcinogens in the body pose to Americans? We can get a glimpse by applying BEs, which were available for nine carcinogens in this report, to exposure levels in the NHANES surveys (Figure 2).


Chemical carcinogens with available BE information:

  • acrylamide, used in the production of industrial polymers and produced when starchy foods are cooked at high temperatures
  • arsenic, a naturally occurring metal
  • benzene, a component of petroleum
  • bromodichloromethane, bromoform and dibromochloromethane, by-products produced when chlorine, used to disinfect water, reacts with organic matter
  • DDT & DDE, pesticide (banned in the U.S.) and its breakdown product
  • hexachlorobenzene, a fungicide and industrial chemical, also a by-product in the production of chlorinated solvents

We used BEs to approximate the proportion of the U.S. population that may be exposed to certain chemicals at levels that exceed what EPA considers to be an acceptable risk.11

  • More than half of the NHANES population had levels of arsenic and acrylamide that were estimated to exceed a 1-in-10,000 cancer risk. More than 10 percent had levels of benzene that exceeded that degree of risk.
  • The majority of Americans tested also had levels of DDT and DDE,* benzene, hexachlorobenzene, bromodichlorobenzene and dibromochloromethane that exceeded the one-in-a-million risk threshold.
  • Only for bromoform were estimated cancer risks for the general population less than one in a million.

Concentrations of these carcinogens did not vary greatly between older and younger people, indicating that exposures and risk for these chemicals start accruing early.

Although biomonitoring equivalents are useful they should be interpreted with caution.
They assume a measured biomarker concentration represents a person's average lifetime exposure of an individual. This may be a reasonable assumption for chemicals that are persistent or chemicals to which we are continuously exposed, but not for chemicals that quickly move through the body, are easily metabolized or have intermittent exposures. Those require repeated samples over time to tell the story of a person’s lifetime exposure. Risk is also highly dependent on genetics and individual differences in metabolism. Finally, BE data was available for only a few select chemicals that may not be representative of risks for other carcinogens.

What the estimates do provide is an indication that at least a small subset of the hundreds of carcinogens measured in humans are contributing to non-trivial increases in cancer risk. These risks may be even greater when we consider that the estimates are only for individual chemicals and do not account for how combined exposures to multiple chemicals may increase risk.

* DDT exposure is customarily calculated from the sum of the levels of DDT and its breakdown product DDE.