| Abstract: |
Background: Evidence linking female hormones to the development of malignant melanoma has been contradictory. The purpose of this study was to examine the risk of melanoma in relation to exogenous and endogenous hormonal variables in women, including oral contraceptives, replacement oestrogens, pregnancy, and menopause. Methods: Hormonal and reproductive factors were evaluated using data from a personal-interview population-based case-control study of melanoma in women conducted in Connecticut during 1987-1989. Caucasian female incident invasive melanoma cases (n = 308) were confirmed by standardized histopathological review. Caucasian female controls (n = 233) were selected by random digit dialling and frequency-matched on age. Data were analysed using multivariate logistic regression. Results: Ever being pregnant, age at first pregnancy, current use of replacement oestrogens, ever use of oral contraceptives, duration of use of oral contraceptives, and age at first use of oral contraceptives were not associated with melanoma. Among other variables, cases were more than twice as likely as controls to report a single pregnancy lasting > 6 months, but this association lacked a dose-response relationship. Menopause and body mass index were not independently associated with risk of melanoma. However, this analysis did suggest that menopause and body mass index may be interactive risk factors. Melanoma cases were three times more likely than controls to be obese and report natural menopause when compared to thin/acceptable premenopausal women (OR = 3.00, 95% CI: 1.03-8.73). Conclusions: These data do not provide strong evidence that hormonal and reproductive factors are associated with risk of melanoma in women, although the few positive results should be explored further. Evidence linking female hormones to the development of malignant melanoma has been contradictory. A population-based case-control study conducted in Connecticut (US) assessed this relationship again. Hormonal and reproductive factors ascertained through personal interviews were compared between 308 Caucasian women diagnosed with incident invasive melanoma in 1987-89 (listed in hospital tumor registry logs) and 233 age-matched controls selected through random digit dialing. Melanoma was not associated with ever being pregnant, age at first pregnancy, current use of replacement estrogens, ever use of oral contraceptives (OCs), duration of OC use, or age at first use of OCs. Cases were more than twice as likely as controls to report a single pregnancy lasting more than 6 months, but this association lacked a dose-response relationship. Although menopause and body mass index were not independently associated with melanoma risk, they may be interactive risk factors. Finally, women with melanoma were three times more likely than controls to be obese and to report natural menopause. These findings suggest that neither exogenous nor endogenous hormones contribute significantly to an increased risk of melanoma. However, the positive associations observed in this study, including the potential effect of hormones on postmenopausal women, merit further investigation. |
