Contrary to the non-steroidal compounds, steroidal compounds like exemestane exhibit androgen agonistic activities in vivo75 expressed as a dose-dependent suppression of sex hormone binding globulin (SHBG). This effect on SHBG is not related to estrogen suppression per se, as it is not observed among the non-steroidal compounds. Notably, it is not caused by exemestane itself but through its main in vivo metabolite 17hydroexemestane.76 Interestingly, when looking at formestane, this compound does not express androgen agonistic activities on sex hormone binding globulin when administered by the parenteral route; it is, however, observed after oral administration, probably due to the excessive first pass effect on the liver.[74] and [77]
While the androgen agonistic activity of exemestane is modest, it may be of clinical importance. Most breast cancers contain androgen receptors above 10 fmol per mg protein,78 and overexpression of androgen receptors as well as stimulation of MCF-7 cells with dihydrotestosterone counteracts estradiol stimulation.79 Androgens were used for breast cancer therapy before the contemporary era of tamoxifen use80 H.B. Nevinny, C.R. Haines, M.M. Dederick and T.C. Hall, Comparative study of 6-dehydro-17alpha-methyltestosterone + testosterone propionate in human breast cancer, Cancer 17 (1) (1964), p. 95.80, and the finding in experimental systems that estrogen deprivation sensitizes breast cancer cells to androgen growth inhibition[81] and [82] raises the possibility that patients under treatment with aromatase inhibitors may be sensitized to the anti-tumour effects of androgens.
In contrast, those studies applying formestane second to a non-steroidal compound each administered formestane parenterally. However, a lack of a first-pass effect on liver synthesis of SHBG may not exclude an androgen-agonistic effect on tumour tissue.