Risk of malignant melanoma in relation to drug intake, alcohol, smoking and hormonal factors J Westerdahl', H Olsson2, A Masback3, C Ingvar' and N Jonsson3 Departments of 'Surgery, 2Oncology, 3Pathology, University Hospital, S-22J85 Lund, Sweden.Summary In a population-based, matched case-control study from southern Sweden of 400 patients with a first diagnosis of malignant melanoma and 640 healthy control subjects aged 15-75 years, the association between commonly prescribed drugs, alcohol, smoking and malignant melanoma was evaluated. In addition, the relation between reproductive and hormonal factors and melanoma in women was studied. It was found that certain specific types of prescribed drugs, i.e. beta-blockers, hydralazines and benzodiazepines, may increase the risk of melanoma development. However, these associations were diminished, at least for benzodiazepines, after controlling for host factors. As these findings are unconfirmed, and may be due to chance or confounding, further studies are warranted. The risk of malignant melanoma was not influenced by alcohol consumption or smoking habits. Our results do not suggest an association between oral contraceptives and melanoma. Furthermore, reproductive factors were not independent risk factors for melanoma. However, increasing number of live births seemed to be protective (P for trend = 0.01). There is a need for further research to be able to draw firm conclusions on the relation between number of live births and melanoma. The results based on histopathological re-examinations and those based on tumour registry data were essentially the same.

In a population-based, matched case-control study from the South Swedish Health Care Region, which has the highest risk for melanoma in Sweden, the relation between the use of sunbeds or sunlamps and malignant melanoma was investigated. Between July 1, 1988, and June 30, 1990, a total of 400 melanoma patients and 640 healthy controls aged 15-75 years answered a comprehensive questionnaire containing different epidemiologic variables. Questions regarding the use of sunbeds or sunlamps were included. The odds ratio for developing malignant melanoma after ever having used sunbeds or sunlamps was 1.3. Considering all age groups, the odds ratio was significantly elevated after exposure more than 10 times a year to sunbeds or sunlamps (odds ratio (OR) = 1.8). When the study was restricted to patients and controls younger than age 30 years because the use of tanning devices is much more common among young persons, the odds ratio was higher (OR = 7.7 for more than 10 times a year vs. none). These findings were independent of constitutional factors and factors regarding sun exposure. A dose-response relation was evident. Furthermore, among melanoma patients in this young age group, the ratio of females to males was significantly higher than in older patients. When different melanoma presentation sites were considered, only lesions of the trunk were significantly associated with sunbed or sunlamp use (OR = 4.2 for more than 10 times a year vs. none).

No abstract

A population-based study from Sweden identified 711 patients with cutaneous malignant melanoma diagnosed in 1965, 1975, 1985 and 1989. Prognostic factors were evaluated and a review of the literature was performed. On univariate analysis, thick tumours (> 0.8 mm) (odds ratio [OR] 1.8, 95% confidence interval [CI] 1.6-2.1), increasing Clark level (OR 1.8, 95% CI 1.6-2.0), ulceration (OR 1.8, 95% CI 1.6-2.0), nodular melanoma (OR 1.5, 95% CI 1.3-1.6) and increasing age (continuous variable, P < 0.0001) were associated with a shorter survival. Location on extremities (OR 0.8, 95% CI 0.7-0.9), inflammation (OR 0.8, 95% CI 0.7-0.9) and female gender (OR 0.8, 95% CI 0.8-0.9) were associated with improved survival. On multivariate analysis, thick tumours (> 0.8 mm) (OR 1.5, 95% CI 1.2-1.7) and ulceration (OR 1.4, 95% CI 1.2-1.6) were independently related to a poor prognosis, while location on extremities (OR 0.8, 95% CI 0.7-0.9), inflammation (OR 0.8, 95% CI 0.7-0.9) and female gender (OR 0.8, 95% CI 0.8-1.0) were associated with improved survival. No difference in mean tumour thickness was seen over time, but there was a significant increase in the percentage of thin melanomas (< 0.8 mm) in 1985 (P = 0.01) and 1989 (P = 0.002) compared with 1965. The incidence of melanomas with inflammation increased significantly (P = 0.04), as did age at diagnosis (P = 0.005).

No abstract