Many human skin tumors contain mutated p53 genes that probably result from UV exposure. To investipte the link between UV exposure and p53 gene mutation, we developed two methods to detect presumptive UV-specific p53 gene mutations in UV-exposed normal skin. The methods are based on mutant allele-specific PCRs and ligase chain reactions and designed to detect CC to TT mutations at codons 245 and 247/248, using 10 jug of DNA samples. These specific mutations in the p53 gene have been reported in skin tumors. CC to TT mutations in the p53 gene were detected in cultured human skin cells only after UV irradiation, and the mutation frequency increased with increasing UV dose. Seventeen of 23 samples of normal skin from sun-exposed sites (74%) on Australian skin cancer patients contained CC to TT mutations in one or both of codons 245 and 247/248 of the p53 gene, and only 1 of 20 samples from non-sun-exposed sites (5%) harbored the mutation. None of 15 biopsies of normal skin from non-sun-exposed or intermittently exposed sites on volunteers living in France carried such mutations. Our results suggest that specific p53 gene mutations associated with human skin cancer are induced in normal skin by solar UV radiation. Measurement of these mutations may be useful as a biologically relevant measure of UV exposure in humans and as a possible predictor of risk for skin cancer.

No abstract

We conducted a case‐control study of squamous cell carcinoma of the skin (SCC) in a cohort of people followed from 1987 to 1994. Subjects were residents of Geraldton, Western Australia, who were between 40 and 64 years of age in 1987. On 2 occasions, in 1987 and 1992, dermatologists examined participants for skin cancers. Subjects were also asked on several occasions about skin cancers that they had had treated. Migrants to Australia had reduced risks of SCC. Furthermore, people who migrated to Australia early in life or, equivalently, lived in Australia for a long time had a higher risk than immigrants who arrived later in life or more recently. People who had southern European ancestry had a much lower risk of SCC than other subjects, most of whom were of British or northern European origin. Among Australian‐born subjects of British or northern European ancestry, the skin's sensitivity to sunlight was strongly associated with SCC. The pigmentary traits of hair colour, eye colour and skin colour showed weaker associations. The degree of freckling on the arm was strongly predictive of risk. The risk of SCC increased strongly with increasing evidence of cutaneous solar damage and was most strongly associated with the number of solar keratoses. Our results show that sensitivity to sunlight and high levels of exposure to sunlight are important determinants of the risk of SCC. Int. J. Cancer 76:628–634, 1998.© 1998 Wiley‐Liss, Inc.

A survey of the incidence and prevalence of non‐melanocytic skin cancer in Geraldton, Western Australia, was undertaken in November 1987. All residents aged 40 to 64 years whose names were on the electoral roll on August 1, 1987 were invited to undergo a whole‐body skin examination by a dermatologist. When a skin cancer was suspected, participants were referred for treatment to their usual medical practitioner. Subjects were asked to recall incident skin cancers over the preceding two years, and medical records were searched for confirmatory evidence. Histological confirmation of all lesions, both prevalent and incident, was sought and sections were obtained for a standardized review. The prevalence of confirmed non‐melanocytic skin cancer in those aged 40 to 64 years was 7.0% in men and 4.7% in women. The prevalence of basal‐cell carcinoma (BCO was 6.5% in men and 4.5% in women while the prevalence of squamous‐cell carcinoma (SCO was 1.2% in men and 0.3% in women. The estimated incidence rate of non‐melanocytic skin cancer in this age group was 1560 per 100 000 person‐years. The estimated incidence rate of BCC in men was 1335 per 100 000 person‐years, and in women 817 per 100 000, while in men the estimated incidence rate of SCC was 890 per 100 000 person‐years, and in women it was 289 per 100 000 person‐years.

To measure the rate at which non‐melanocytic skin cancers develop, we conducted a population‐based, longitudinal study in Geraldton, Western Australia. Subjects were residents of Geraldton, Western Australia, who were between 40 and 64 years of age and registered on the electoral roll in 1987. In 1987 and again in 1992, dermatologists examined participants for skin cancers. They examined all skin areas, apart from those covered by underwear or hair. Subjects were asked about skin cancers that they had had treated between the 2 surveys. When all skin cancers were counted, the incidence rates of basal cell carcinoma were 3,379 per 100,000 person–years in women and 7,067 per 100,000 in men; those of squamous cell carcinoma were 501 per 100,000 in women and 775 per 100,000 in men. Sixteen percent of men and 14% of women developed at least one basal cell carcinoma; 2.8% of men and 2.2% of women had at least one squamous cell carcinoma. Most incident skin cancers were diagnosed at the second examination. More than half of the subjects who had a skin cancer at the first examination developed another. Squamous cell carcinomas occurred almost exclusively on parts of the body that are usually exposed. Basal cell carcinomas were common on the head, neck and trunk but not on the forearms and backs of hands. A quarter of people with a skin cancer on an exposed site also had one on the trunk. Our results show that skin cancer is extremely common in this population and frequently undiagnosed. Multiple skin cancers occur commonly, and skin cancers on exposed sites often are associated with skin cancers on less exposed sites. Int. J. Cancer 73:629–633, 1997. © 1997 Wiley‐Liss, Inc.