Does endometriosis increase a women's melanoma risk?

Journal of Drugs in Dermatology, Jan, 2008 by Kendra Gail Bergstrom

Abstract

Case-control and cohort studies have suggested an association between melanoma and endometriosis. A prospective study of over 90,000 patients in the French national health care system, is the largest study to evaluate the question. A history of endometriosis or uterine fibroma was associated with an increased risk of melanoma (relative risk, 1.62; confidence interval 1.15-2.29 for endometriosis). No association was seen between melanoma and ovarian cysts, uterine polyps, breast adenoma, or fibrocystic disease of the breast. The association between melanoma and endometriosis may be due to hormonal influences or shared genetic mutations between the 2 diseases.

Endometriosis is a common disease in women, with estimated prevalence of 5%-10% of the female population. For patients with dymenorrhea or infertility undergoing evaluation, the prevalence is 60% and 30% respectively. The cause of endometriosis is not clear and hypotheses range from retrograde menstruation, metaplastic conversion of normal epithelium to endometrial cells, hormonal imbalances, and local allelic imbalance of tumor suppressor genes (1,2) such as pl6INK4A or CDKN2A, p53, and PTEN.

This study in the Archives of Internal Medicine in October 2007 (3) prospectively followed a cohort of 98,995 teachers in France covered by a national health care system for a median of 12 years between 1989 and 2002. Subjects were questioned by survey at the beginning of the study and every 2 years. Approximately 5949 patients (6%) had a history of endometriosis at the end of the study. Endometriosis was diagnosed when a patient had either a diagnostic or therapeutic procedure.

Three hundred and sixty three melanoma cases were reported over the 12 year period. Melanoma cases were confirmed by pathology reports in most cases (98%), and the remainder by confirmation of a primary physician. No distinction was made between in situ and invasive melanoma, and no pathologic review was performed. Not surprisingly, women with melanoma were more likely to have fair skin, sun sensitivity, large numbers of nevi and freckles, and blond, red, or chestnut hair.

A personal history of endometriosis significantly increased the risk of melanoma in women. The relative risk of melanoma for a woman with endometriosis was 1.62 (95% confidence interval 1.15-2.29). The study followed other gynecologic diseases, including ovarian cysts, uterine fibromas, uterine polyps, breast adenoma, and fibrocystic changes. The only other positive association was that with uterine fibromas, with a relative risk of 1.33 (95% confidence interval 1.06-1.67). Uterine fibromas were by far the most common gynecologic tumor, present in almost 25% of patients.

Is there an association between endometriosis and cancer?

Studies over the past 20 years have yielded conflicting results as to whether endometriosis is associated with increased cancer risk. Because of the proliferative state associated with endometriosis, several studies have evaluated its association with different cancers. Observational, case-control, and cohort studies show an increased risk of breast cancer and ovarian carcinoma. (4,5) The relative risk of these malignancies is between 1.3 and 1.9, similar to the risk of melanoma seen in this study. Previous smaller studies have suggested a link between endometriosis and melanoma. Due to their smaller size or low number of melanoma cases, previous overall findings were not significant. Of interest to dermatology, studies have observed associations between endometriosis, dysplastic nevi, melanoma, and red hair color. (6,7,8)

Why would a melanoma risk be increased by endometriosis?

Estrogen receptors of both alpha and beta subtypes are found in both benign and malignant melanocytic lesions. (9) Hyper-methylation of estrogen receptor alpha, as measured in human serum, is associated with melanoma progression. (10) Previous studies of tamoxifen, a selective estrogen receptor modulator, showed initial promise in melanoma, though the addition of tamoxifen did not increase survival in a subsequent Phase III clinical trial. (9)

Pregnancy is another hyperestrogenic state associated with increased proliferation of melanoma. Melanomas diagnosed during pregnancy are usually thicker in depth, (10) though the outcome and survival data are comparable to nonpregnant patients with melanoma of the same depth. (10) While the etiology of the thicker melanomas is not clear, the presence of estrogen and progesterone receptors on tumor cells suggest a role for hormonal influences.

Therapeutic implications

The association of melanoma with endometriosis suggests a role for hormonal influences and a possible therapeutic target in estrogen receptor modulation. While trials in cutaneous melanoma did not show benefit over a large group, (11) clinical trials are now underway studying the use of selective estrogen receptor modulators such as tamoxifen in ocular and esophageal melanoma. It is possible that an undetermined subset of melanomas is sensitive to estrogen modulation.