Melanoma, the most deadly form of skin cancer, has tripled in incidence since 1975 in the USA. The World Health Organization estimates that in 2000, there were over 200,000 cases of melanoma diagnosed and 65,000 melanoma-associated deaths worldwide.¹

The main risk factors for the disease are exposure to ultraviolet radiation and having fair skin, as well as age and genetic factors. Although other factors, such as exposure to chemicals, have been studied, no consistent associations have been found.

A study published this week by a team led by Leslie Dennis, of the University of Iowa, USA, links exposure to certain pesticides in farm workers to an increased risk of melanoma.

“Most previous melanoma literature has focused on host factors and sun exposure, but our study suggests more research is needed on chemicals and other environmental factors,” they write in Environmental Health Perspectives.²

The researchers analysed data on melanoma incidence in over 50,000 workers in two US states, Iowa and North Carolina, who had ever applied any one of 50 pesticides. The data were taken from the Agricultural Health Study, an ongoing epidemiological study carried out by the US Environmental Protection Agency and the National Institutes of Health that began in 1993. The authors were able to estimate the dose of each pesticide by estimating the cumulative exposure in days over a lifetime, and adjusting for application method and use of protective equipment.

After adjusting for known risk factors such as sun sensitivity and exposure, they found a significant dose-response relationship between incidence of melanoma and having applied each of four pesticides: the insecticides parathion and carbaryl, and the fungicides mancozeb and benomyl.

They also found a link, but without a dose-response, between the cancer and inorganic arsenic-containing pesticides, although only 44 participants reported using these. In addition, lead arsenate exposure significantly increased the strength of the association between cancer and the two fungicides, implying that the two chemicals act synergistically.

Because the authors studied 50 pesticides, the results should be interpreted with caution, warns Paolo Boffetta, of the Mount Sinai Medical School in New York City, USA. Comparing multiple possible risk factors can increase the chance of a false-positive result, he explains.

Nevertheless the results for two of the pesticides, mancozeb and parathion, were highly significant, Boffetta points out, and so are unlikely to be false positives.

“Multiple comparisons may be an issue,” the authors acknowledge. Because of this, they focussed on associations at the 1% significance level before adjusting for confounders, and considered existing evidence for biological mechanisms in interpreting their results, they claim. “These results should also be interpreted with regard to their consistency with other studies,” they write.

Although this is a strong study, Boffetta says, he points out that melanoma is still a relatively rare cancer and the total number of cases included in the study was only 271, despite the high numbers of participants. In addition, pesticide applicators are exposed to a cocktail of other chemicals, which may interact in inducing any carcinogenic effects.

Several analyses of data from the Agricultural Health Study are showing signs of a link between pesticides and a number of other cancers, Boffetta says, and this is a reason for concern. While this study adds to that body of evidence, “it’s frustrating not to have an overview,” he says.

The link between pesticides and melanoma should also be examined in the general population, suggests Boffetta, as parathion and other pesticides have been widely used in household products.