Authors affirm that estimating attributable risks from carcinogens is based on “unverified assumptions” in which evidence, some strong and some weak, is treated equally. Effective primary cancer prevention should reduce carcinogen numbers or exposures. Experimental evidence with mechanisms of action and suggestive epidemiology linking cancers and exposures is likely to produce a larger and better list of human carcinogens and target organs.  The paper notes: “Effective primary prevention resulting in a reduction of cancer risk can be obtained by: (i) a reduction in the number of carcinogens to which humans are exposed; or (ii) a reduction of the exposure levels to carcinogens. Exposure levels that could be seen as sufficiently low when based on single agents, may actually not be safe in the context of the many other concomitant carcinogenic and mutagenic exposures.”

L Tomatis, J Huff, I Hertz-Picciotto and others. Avoided and avoidable risks of cancer, Carcinogenesis, number 18, number 1, pages 97–105, 1997.