Valid evidence from randomized controlled trials (RCTs) indicates that breast cancer risk is increased with combined estrogen/progestogen use and that such treatment implies a risk beyond that of estrogen alone. Available evidence from observational studies, including nearly two million women, broadly agrees with trial findings. Overall, risk estimates from observational studies are somewhat higher than in RCTs but remain modest compared with other risk factors even after long-term treatment. For combined estrogen/progestogen therapy, risk is gradually increasing to become significant after 4–5 years. After cessation of therapy, risk estimates decline to reach baseline levels after around 5 years. Whether the results are valid for all progestogens as a group or if there are indeed alternative progestogen compounds without stimulatory effects on the breast remains to be elucidated. It seems reasonable that breast cancer risk should depend upon the dosage of hormones given, but so far information on this point is insufficient and needs further study.
The many questions about breast safety are of major importance for all compounds to be used in menopausal hormone therapy. Apart from its many beneficial health effects, for estrogen alone also the safety data are quite reassuring. The only justifications for progestogen addition are bleeding control and endometrial protection. At present, there are several new therapeutic compounds and principles in development which hold promise to provide both endometrial protection and breast safety. The administration of ultra-low doses of oral or intrauterine progestogen in combination with estrogen has been shown to have only little effect on breast density. Anti-progestogens and androgenic compounds may not stimulate breast cell proliferation. On theoretical grounds, combinations of estrogen together with different selective estrogen receptor modulators should also be able to avoid any stimulation of the breast.
For apparent reasons, prospective long-term large-scale clinical trials with breast cancer as the primary endpoint are not feasible. However, much important knowledge can still be obtained from short-term studies using relevant surrogate markers for breast cancer risk. During recent years, accumulating data from animal, clinical and observational studies strongly suggest a proliferative effect of progestogens in human breast tissue. Findings in surrogate markers like breast density and proliferation add to epidemiological reports and randomized trials. From a clinical perspective, increased breast epithelial proliferation and mammographic density during hormonal treatment should be regarded as unwanted and potentially hazardous side-effects. Efforts should be made to identify those women who are most sensitive and react with excessive proliferation during treatment. Still, in many women no apparent increase in density or proliferation will be detected. Not all women respond in the same way to the same treatment and the biological basis for the marked individual variation has to be clarified. There is an apparent need to develop clinical and non-invasive methods for monitoring breast response during hormonal treatment in order to identify women at risk for breast cancer. Prospective randomized clinical studies are needed to explore the effects of different compounds, doses and regimens on normal breast cell homeostasis.