Reference
Fournier A, Mesrine S, Boutron-Ruault M-C, Clavel-Chapelon F. Estrogen-progestagen menopausal hormone therapy and breast cancer: Does delay from menopause onset to treatment initiation influence risks? J Clin Oncol. 2009;27:1-7.
Design
Prospective case-cohort study
Participants
98,995 French women (in France) who were born between 1925 and 1950
Study Parameters Assessed
At study entry (baseline), participants completed questionnaires regarding menopause history, use of menopausal hormone therapy, other lifestyle behaviors, and cancer occurrence. Cancer outcomes were assessed from annual self-administered questionnaires with 93% of cancers confirmed with a review of pathology reports.
Key Findings
Estrogen-progestagen (levonorgestrel) ever-users were at significantly increased breast cancer risk compared with hormone therapy never-users. Women who had used estrogen-progestagen hormone therapy within 3 years of the onset of menopause had a 61% increased risk for developing breast cancer. Women who started estrogen-progestagen hormone therapy more than 3 years after the onset of menopause had a 35% increased risk of breast cancer, although this finding did not reach statistical significance. Even short-duration hormone therapy conferred increased risk of breast cancer. Women who used estrogen-progestagen therapy for a short duration (<2 y) within 3 years of menopause had a 54% increased risk of developing breast cancer. Conversely, there was no increased risk of breast cancer in women who used estrogen-progestagen hormone therapy for a short duration when this therapy was initiated more than 3 years after menopause. Furthermore, the type of progestin used altered the risk of breast cancer. Short-duration hormone therapies of estrogen-progesterone initiated close to menopause did not increase the risk of breast cancer. Estrogen and dydrogesterone conveyed intermediate risk of breast cancer development and estrogen plus other progestagens doubled the risk of breast cancer. Estradiol was the form of estrogen in all cases of hormone therapy.
Practice Implications
This study adds important clarifying data to the body of evidence supporting a link, albeit a complicated one, between menopausal hormonal therapy and breast cancer. Essentially, initiation of menopausal hormone therapy close to menopause onset, as opposed to later, still results in an increased risk of breast cancer. Even short duration (less than 2 y) of use of estrogen-progestagen therapy initiated within 3 years of menopause is associated with significantly increased risk of breast cancer. The only exception to this increased risk was hormonal therapy that specifically included progesterone (as opposed to other progestins), indicating that the form of progestagen is important to cancer risk as different forms of progestagens may affect breast cancer cells differently. All forms of estrogen-progesterone hormone therapy taken for longer durations (>2 y) increased the risk of breast cancer. Thus, long-term use of estrogen and any form of progesterone conferred an increased risk of developing breast cancer.
While this study did not address the risk associated with bioidentical hormone therapy, it does caution practitioners away from the notion that short-term hormone replacement therapy initiated soon after menopause is safe. In fact, short-term estrogen and progestagen given soon after the initiation of menopause double the risk of breast cancer. The question of whether any hormonal therapy—bioidentical or not—poses an increased risk of breast cancer risk or whether the type of hormones modify this risk remains essentially unanswered. However, this study hints at the fact that the type of hormone used might, in fact, make a difference.
Limitations
This study was limited by the reliance upon self-reported data and thus the possibility of recall bias. However, the biennial schedule of questionnaires likely limited this bias. The relatively small number of breast cancer cases may have limited the ability to detect significant associations or variations in the observed hazard ratios. In general, the large study population that included good representation of all subgroups gives this study validity.
