Pigmentary and cutaneous risk factors for non-melanocytic skin cancer--a case-control study.

The roles of ethnic origin, pigmentary traits, sun sensitivity and other cutaneous characteristics as risk factors for basal-cell carcinoma (BCC) and squamous-cell carcinoma (SCC) were examined in a case-control study of prevalent and incident cases of histopathologically confirmed skin cancers. Two hundred and twenty six confirmed cases of BCC, 45 of SCC and 1,015 controls with no lesions were identified in a population-based survey of skin cancer in 1987 in Geraldton, Western Australia. The risk of both cancers was higher in native-born Australians than in migrants. The risk of BCC decreased with increasing age at arrival in Australia. Southern European ancestry was strongly protective against BCC (for any southern European grandparents) and SCC (no case of SCC had any grandparents of southern European origin). Inability to tan was the strongest pigmentary risk factor for both BCC and SCC. Among factors that incorporated a measure of sun exposure as well as sun sensitivity, freckling on the arm in childhood was important for both cancers, the number of moles on the back was important for BCC, and forearm skin colour and having a permanent colour difference between the neck and adjacent protected areas were important for SCC. Among measures of sun damage to the skin, solar elastosis of the neck was a strong risk factor for both BCC and SCC, loss of fine texture of the skin of the back of the hands (as measured by cutaneous microtopography) was important for BCC and telangiectasia of the face for SCC. When all important variables for each cancer were examined together in a single model with age, sex, migrant status or age at arrival in Australia, and ethnicity, in ability to tan, solar elastosis of the neck, and the number of moles on the back were independently significant risk factors for BCC and solar elastosis of the neck and having a permanent colour difference between the neck and adjacent protected areas were independently significant risk factors for SCC. The effects of age at arrival or migrant status and ethnic origin remained important in the models incorporating these factors. A history of ever having acne and a history of warts were protective for BCC and a history of acne was protective for SCC.