Nonmelanomatous skin cancer following cervical, vaginal, and vulvar neoplasms: etiologic association.

Human papillomavirus infection is the major cause of cancers of the cervix, vagina, and vulva (1). Nonmelanomatous skin cancers have been associated with human papillomavirus infection in patients with epidermodysplasia verruciformis and in patients who are immunosuppressed or nonimmunosuppressed, although the data are scant (1,2). We used the cancer registry of the Swiss Canton of Vaud (with a population of approximately 600 000 in 1990) for the period from 1974 through 1994 to obtain additional quantitative information on this topic, which has pathogenic and public health implications. Data were collected for women who had in situ or invasive neoplasms of the cervix, vagina, or vulva and for women who had nonmelanomatous skin cancer. These data were then used to calculate the incidence of nonmelanomatous skin cancer in women who had been registered with an in situ or invasive neoplasm of the cervix, vagina, or vulva (3). The registry is tumor based, and multiple primary tumors in the same person are entered separately. The basic information available consists of sociodemographic characteristics of the patient, the primary site of the tumor, the histologic type of the tumor according to the standard International Classification of Diseases (ICD) for Oncology (4), and the time of diagnostic confirmation. Passive and active follow-ups are recorded, and each subsequent item of information concerning a registered cancer is used to complete the record of that patient. Since 1974, a registration scheme that applies the standardized rules used for incident cancers has been used for carcinoma in situ and severe dysplasia (CIN III, cervical intraepithelial neoplasia III) of the uterine cervix (ICD code: 180.0–180.9), vagina (ICD: 184.0), and vulva (ICD: 184.1–184.3) (4). In the present study, when all synchronous neoplasms were excluded, there were 2339 histologically confirmed cases of carcinoma in situ of the cervix uteri, nine cases of carcinoma in situ of the vagina, and 85 cases of carcinoma in situ of the vulva. The study also included 789 cases of invasive neoplasms of the cervix, 69 cases of invasive neoplasms of the vagina, and 153 cases of invasive neoplasms of the vulva. These cases were followed to the end of 1996 for the occurrence of cancer, migration, or death. We calculated the expected numbers of individuals with nonmelanomatous skin cancer based on site-, age-, and calendar-period-specific incidence rates, multiplied by the observed number of person-years at risk. The statistical significance of the observed/expected ratios (standardized incidence ratio [SIR]) and the corresponding 95% confidence interval (CI) were based on the Poisson distribution. Table 1 gives the observed and expected numbers of nonmelanomatous skin neoplasms after diagnosis of in situ or invasive neoplasms of the cervix, vagina, and vulva. A statistically significant excess of skin cancer was registered after cervical neoplasms (44 observed and 24 expected; SIR 4 1.8; 95% CI 4 1.3–2.5) and vulvar neoplasms (13 observed and four expected; SIR 4 3.2; 95% CI 4 1.7–5.5). Likewise, three nonmelanomatous skin cancers were observed after vaginal neoplasms versus one expected (SIR 4 2.9; 95% CI 4 0.6–8.6). Overall, 60 skin cancers were observed versus 29 expected (SIR 4 2.1; 95% CI 4 1.6–2.7). An excess of nonmelanomatous skin cancer after diagnosis of carcinoma in situ of the cervix has been reported (5,6). The present data extend this observation to other neoplasms of the lower female genital tract and, therefore, provide epidemiologic support to the suggestion of a possible role of human papillomavirus infection in the etiology of nonmelanomatous skin cancer (7).