Carcinogens off the Shelf?

Should you be concerned that a product on the shelf in your local supermarket or drugstore may cause cancer? Some people are. In this issue of the Journal, Jacob et al. (1) have accessed cancer-related reports concerning cosmetic products on file with the Adverse Event Reporting System within the US Food and Drug Administration’s Center for Food Safety and Applied Nutrition. They note from the outset the impact of media coverage on community awareness and that the majority of associations made between product usage and cancer incidence currently lack conclusive evidence. But even if the evidence is equivocal at best in most cases, the burden of anxiety born by many people is evident from this study. Can that burden be lifted, or at least reoriented with reference to proven causes of cancer, through initiatives by cancer professionals? Anxiety about insidious causes of cancer is not focused on cosmetics. From pesticide and packaging residues in food through to electromagnetic fields generated by mobile phones, there has been both research and media attention. Cancer is the most feared disease, and the medico-scientific sector, the politico-legal sector, and the media can all be recognized as having striven to protect the general community from exposure to carcinogens. Recognition that there can be no safe level of exposure to a known carcinogen has served to reduce the incidence of workrelated cancers, often belatedly and mainly in high-income countries. On the matter of particular chemicals in consumer products, however, reduced cancer incidence has not been recognized, mainly because no burden of attributable malignant disease has been established. The term “consumer products” as used in this Editorial excludes tobacco products and alcoholic beverages. Tabulation of the proportion of cancer attributable to particular risk factors in high-income countries identifies the usual suspects: tobacco smoking, diet and obesity, alcohol drinking, sun exposure, infections, pharmaceutical drugs, occupation, and pollution (2). The category “consumer products” is never mentioned. Confidence that the burden of cancer due to consumer products is, at worst, very low, however, does not negate the consideration that even a single case of such cancer is an outrage to the community. Most commonly, specific chemical carcinogens have been identified or confirmed as causing cancer in humans by studies of particular workers (3). In addition to proven carcinogens, there are likely carcinogens: chemicals that cause cancer in experimental animals and for which epidemiological data are equivocal or nonexistent. Through using some consumer products, people may be exposed to levels of particular carcinogens that most commonly are orders of magnitude less than occurred in the circumstances identifying or implicating their carcinogenicity. Even so, under the “no safe dose” argument presented above, such exposure obliges manufacturers to reduce levels, with this imperative being supported by regulation in many contexts (4). Even when evidence of carcinogen exposure is unequivocal, and specifically exposure that has occurred as a result of using a particular consumer product, a consequential burden of cancer in the population that has used the product in question cannot be presumed. There may be no increased incidence, or an increased incidence that is so small as not to be evident using currently available epidemiological procedures. Such an understanding may emerge when reference is made to the level of exposure that using the product in question has entailed by comparison with the levels of exposure documented in other contexts that have been the basis of epidemiological or experimental studies of that carcinogen. Yet media notification of exposure often includes specific comment on the likelihood of cancer causation to engage public attention. For a very limited number of consumer products, specifically including cosmetics, epidemiological data involving consumer usage are available. Primary among these, as Jacob et al. specifically address, are hair dyes. In the 1970s, experimental data inferring a carcinogenic risk prompted replacement of simple aromatic diamines in semipermanent dyes (5). Available epidemiological data concerning personal use of hair colorants fell short of a “possibly carcinogenic to humans” evaluation in a recent International Agency for Research on Cancer (IARC)