Sexual hormones share a common cholesterol-derived steroid backbone and comprise oestrogens, gestagens and androgens. They elicit numerous physiological and pathological functions; for example androgens are known to contribute to prostatic cancer development, whereas oestrogens are of major importance in the pathogenesis of breast cancer. In 1986, the presence of oestrogen receptors (ER) on breast cancer cells was described as one of the first biomarkers in oncology.1 ER expression is still routinely used both for predictive and therapeutic purposes with the selective ER modulator tamoxifen remaining an efficacious breast cancer treatment both in the adjuvant and palliative setting. The ‘classical’ ERs (ERα and ERβ) are found in many other tissues, such as lung, colorectal mucosa, thyroid tissue or lymphoid cells. Observations of darkening or sometimes growing melanocytic nevi in pregnancy as well as of prognostic differences between female and male melanoma patients suggested that sexual hormones may also impact on melanocyte function and melanoma development.2 Both ‘classical’ ERs have been detected in melanoma tissue, with ERβ being the predominant subtype 3 Already over 20 years ago, tamoxifen was tested in combination with chemotherapy in the treatment of metastatic melanoma and demonstrated improved response rates compared to chemotherapy alone.4 However, a meta-analysis of subsequent clinical trials with tamoxifen did not detect a positive influence of additional tamoxifen treatment on overall survival in advanced melanoma.5 The pathogenetic impact of ER expression differs between different cancer types. For example, ERβ proves to have an antiproliferative effect in prostate cancer, while, in contrast, its expression is associated with poor prognosis in ovarian cancer.6, 7 Besides ER α and β, also ‘nonclassical’ oestrogen receptors such as the G protein-coupled oestrogen receptor (GPER) are investigated with growing interest. In the present issue of this journal, Fabian et al.8 report on differential expression of GPER in a total of 81 primary melanomas, either pregnancy-associated melanoma (PAM) or nonpregnancy-associated melanoma (NPAM). GPER is thought to allow for alternative oestrogen-mediated signalling via the MAP-kinase and other intracellular pathways in cells lacking classical ERs. As for other ERs, its role in the pathogenesis of hormone-related cancer is ambiguous with both tumour-promoting and tumour-protective effects proposed in the literature.9 In melanoma, Fabian and colleagues found a significantly higher immunohistochemical and immunofluorescent cytoplasmatic expression of GPER in PAM compared with NPAM (78.4% vs. 27.9%; P = 0.0000), while similar frequencies of ER β expression were reported in both groups (81.6% vs. 60.5%). For immunohistochemical analyses, the authors used ER antibody clones that have recently been considered as specific.10 This is important, because a number of other available ER antibodies have been shown not to stain consistently or to have problems with cross-reactivity. Double receptor expression of ER β and GPER was detected in 73% and 28% of PAM and NPAM, respectively, and was associated with established positive prognostic markers such as lower Breslow thickness, lower mitotic rate and presence of peritumoural lymphocyte infiltration. Hence, the authors suggest that immunohistochemical double ER expression may serve as a novel prognostic indicator in melanoma. This hypothesis needs additional investigation. First, because double receptor expression was not associated with improved disease-free survival in multivariate analysis, which may be attributed to the small sample size of the study, but may also reflect that ERs are predominantly expressed on thinner, less proliferative primary melanomas for other, yet unknown reasons. Second, although ER expression was significantly more frequent in PAM, no survival difference was demonstrated for PAM compared to NPAM, which is in line with the literature on melanoma in pregnancy.11 In addition, the relevance of driver mutations such as BRAF or NRAS for ER expression is not understood. Taken together, the influence of oestrogen and associated receptor expression patterns on melanoma development remains unclear. Further research in this field is needed, especially as the superior prognosis of female melanoma patients continues to imply a role of sexual hormones in melanoma pathogenesis. Contrary parts of oestrogens (tumour-promoting) and progestins (protective) have been recently suggested in this context.12 Moreover, the role of antihormonal therapeutics such as tamoxifen may be worth rethinking in selected melanoma patients.13
Read full abstract