Risk of breast cancer with progestins: critical assessment of current data

Abstract

Whether progestins protect against the risk of breast cancer or enhance that risk has been a major area of controversy over the past several years. Observational studies have reported conflicting results and experimental studies examining whether progestins exert mitogenic or anti-mitogenic actions on breast tissue report divergent results. Based upon a wide range of animal, epidemiologic and clinical data, most investigators agree that estrogens contribute to the development of breast neoplasms. However, the additional effect of progestins on this risk has been the subject of substantial discussion and controversy. A variety of experiments have been carried out using human breast cancer cells grown in vitro and as xenografts in nude mice. These studies demonstrated both mitogenic and anti-mitogenic effects depending upon the precise experimental conditions. Two potential reasons for these differences include differential metabolism of progestins into inhibitory pregnenes or stimulatory 5-alpha-reduced pregnanes or the presence of a protein (GPR 30) which allows the anti-mitogenic effects of progestins to be manifest. Based upon the conflicting nature of the results in experimental studies, we believe that only data in patients provide substantial insight into the actions of progestins on the intact human breast. Studies have now demonstrated that cell proliferation and breast density is higher during the luteal than during the follicular phase of the menstrual cycle. In postmenopausal women, long-term exposure to estrogen plus a progestin results in a marked enhancement of proliferation of the terminal duct lobular units as well as in breast density. These data, taken together, provide substantial evidence that progestins are mitogenic on the human breast when given long term to postmenopausal women. To critically evaluate the observational studies regarding breast cancer risk from progestins, we developed a set of stringent criteria for acceptance of individual studies. Four of the five studies meeting these criteria reported a greater risk of breast cancer with combination estrogen/progestin regimens than with estrogen alone. More importantly, the first randomized, prospective, controlled trial of the risk of breast cancer with an estrogen/progestin combination (the Women’s Health Initiative Study) has now been published. This study reported a 26% increased relative risk of breast cancer with the estrogen/progestin combination. Based upon these data, we believe that progestins do add to the risk of breast cancer over and above that imparted by estrogen alone. The attributable risk during use for 5 years or less is small but increases logarithmically during long-term use. The majority of data regarding progestins are derived from regimens using MPA. However, we conclude from our analysis that the burden of proof regarding progestins has now shifted. One must now prove that an estrogen/progestin combination is safe with respect to breast cancer rather than having to prove it harmful.