Hormonal Management of Menopausal Symptoms in Women With a History of Gynecologic Malignancy

Benjamin S. Harris, MD, MPH; Katherine C. Bishop, MD; Jeffrey A. Kuller, MD; Anne C. Ford, MD; Lisa C. Muasher, MD, MPH; Sarah E. Cantrell, MLIS; Thomas M. Price, MD


Menopause. 2020;27(2):243-248. 

In This Article

Breast Cancer

Systemic Hormone Therapy in Women With a Personal History of Breast Cancer

Breast cancer treatment with aromatase inhibitors (AIs), tamoxifen, therapeutic bilateral salpingo-oophorectomy, or gonadotoxic chemotherapy, may lead to severe, distressing side effects from hypoestrogenism. Up to 95% of breast cancer survivors report vasomotor symptoms, which are more severe than in patients without breast cancer.[1] These women may seek hormonal treatment, making it important to understand the effect of HT on the breast and the risk of cancer recurrence.

A 2005 meta-analysis of two randomized controlled trials (RCTs) and eight observational studies evaluated systemic HT in breast cancer survivors. In the two included RCTs, patients had a mean age of 55.5 years, duration of HT use of 19.9 months, and a mean follow-up period of 25.2 months. Pooled analysis demonstrated a three-fold increased risk of recurrence (relative risk [RR] 3.41, 95% confidence interval [CI] 1.59–7.33). In the observational studies, the mean patient age was 59.7 years, HT use was 28 months, and the mean follow-up period was 57.1 months. Pooled analysis of these studies showed a decreased risk of recurrence (RR 0.64, 95% CI 0.50–0.82). The studies in this meta-analysis included both estrogen plus progestin therapy (EPT) and estrogen therapy (ET) alone, making it difficult to stratify results by hormone type, formulation, dose, or route of delivery, and the observational studies may have been subject to selection, reporting, or publication bias. The authors were unable to explain these conflicting results, and they concluded that by design, the RCTs were more reliable and they therefore would not recommend systemic HT in patients with breast cancer.[2]

In the Hormonal Replacement After Breast Cancer—Is It Safe? (HABITS) trial, a Swedish randomized non–placebo-controlled noninferiority trial aimed at detecting an increased risk of a new breast cancer in breast cancer survivors, HT was associated with significantly more breast cancers at a 2-year interim analysis (hazard ratio [HR] 3.3, 95% CI 1.5–8.1), leading to early termination of the study. The authors continued to follow these patients, and at 4 years after the study began, the risk of breast cancer recurrence remained significantly increased (HR 2.4, 95% CI 1.3–4.2).[3,4] The type of therapy used was predominantly continuous EPT or cyclic estradiol and norethisterone, but a variety of regimens were allowed based on provider preference. Baseline characteristics were similar, except that the HT group had more patients with estrogen receptor–positive cancers (62.3% vs 54.5%). In the 4-year follow-up analysis, the HR of a new breast cancer for women with hormone receptor–positive cancers was 2.6 (95% CI 1.3–5.4) compared to 1.8 (95% CI 0.7–4.8) for women with hormone receptor–negative cancers.

Another Swedish RCT conducted around the same time, known as the Stockholm study, did not show an elevated recurrence risk after 4 years of HT use (HR 0.82, 95% CI 0.35–1.90).[5] Patients on EPT used cyclic estradiol, 2 mg for 21 days, plus medroxyprogesterone acetate (MPA), 10 mg daily for the last 10 days, with 1 week of no treatment, or, if they were older than 55 years, 84 days of estradiol and 14 days of MPA were recommended. Patients with a prior hysterectomy were prescribed estradiol 2 mg daily. Given slow recruitment in the Stockholm study and the early results of the HABITS trial, the two studies combined their data and a joint analysis showed an HR of 1.8 (95% CI 1.03–3.1) for breast cancer recurrence.[5] The analyses of each trial was then updated after a longer follow-up. As mentioned above, the HABITS trial continued to show an increased risk with an HR of 2.4 (95% CI 1.3–4.2), whereas the Stockholm trial demonstrated no increased risk (HR 1.3, 95% CI 0.9–1.9).[3,6] The authors of the Stockholm study were unable to explain the large difference in results, but they noted a higher proportion of their patients on tamoxifen than in the HABITS trial (52% vs 21%, respectively). They also attempted in their study protocol to minimize progestin exposure by placing patients on cyclic therapy given that this method was hypothesized to decrease expression of local growth factors and initiate apoptosis of cancer cells.[5]

A recent systematic review of one RCT and three observational studies examining the recurrence risk in premenopausal breast cancer survivors younger than 50 years using HT concluded that there was no significant association between HT and breast cancer recurrence risk (RR 1.04, 95% CI 0.45–2.41) when pooling the study data. Individually, two of the observational studies showed a lower recurrence risk, and one showed a higher recurrence risk with HT.[7] The only RCT, the LIBERATE trial, used a continuous dose of oral tibolone, a synthetic steroid hormone not available in the United States that acts like estrogen in bone and vaginal tissue, and acts like progesterone in the endometrium. This study showed a significantly increased risk of recurrence in women aged 40 to 49 years (HR 1.56, 95% CI 1.1–2.2) as well as in women aged 50 to 59 years (HR 1.45, 95% CI 1.08–1.9), leading to early termination of the study.[8] The conflicting results in these few studies and overall lack of data regarding the use of HT in premenopausal patients with a history of breast cancer limited the authors from drawing any definitive conclusions.

ET has the potential to reduce the incidence of breast cancer through inducing apoptosis of cancer cells.[9,10] This mechanism would explain why ET in the Women's Health Initiative[11,12] and other trials[13–15] showed a decreased risk of breast cancer, and why initiating EPT or ET after a delay of several years has been shown to have decreased risks of breast cancer as opposed to initiating therapy soon after menopause.[16] According to this hypothesis, estrogen deprivation, as occurs in the first few years of menopause, is needed to allow occult breast cancer cells to become vulnerable to estrogen-induced apoptosis. Without estrogen signaling them to grow, many of these cells die, with a few cells adapting to an estrogen-depleted environment and surviving. When re-exposed to estrogen, apoptosis is triggered. Although some cells may continue to survive, there appears to be a balance of replication and apoptosis to maintain a stable population. This idea could be applied clinically with selective estrogen receptor modulators (SERMs; eg, bazedoxifene) inducing estrogen deprivation followed by ET to induce apoptosis of sensitized cancer cells.[9] Although intriguing, more research is needed with clinical trials in this area before advocating for the use of these products in routine clinical practice.

With the available evidence, The North American Menopause Society (NAMS) posits that systemic HT (ET or EPT) is not recommended in breast cancer survivors except in select cases in which the risks have been discussed in conjunction with an oncologist and nonhormone options have failed.[17]

Low-dose Vaginal Estrogen in Women With a Personal History of Breast Cancer

Low-dose vaginal ET refers to the vaginal ring (Estring), which releases estradiol 7.5 μg/day, Vagifem (Novo Nordisk A/S, Copenhagen, Denmark) tablets (10 μg estradiol given twice weekly, lowered from prior dosages of 25 μg/day), and lower doses of vaginal creams containing estradiol (Estrace; Allergan Pharmaceuticals, Madison, NJ) or conjugated equine estrogen (CEE) (Premarin Vaginal Cream; Wyeth-Ayerst, Philadelphia, PA).[18] There is good evidence that circulating estrogen levels are minimally increased with low-dose vaginal ET.[17,19,20] In the initial 2 weeks of treatment, serum levels are highest, but after this point, serum estradiol levels become similar to levels after natural menopause (<20 pg/mL).[20] In addition, a recent systematic review including all the available low-dose vaginal estrogen products concluded that low-dose unopposed vaginal estrogen does not increase the risk of endometrial hyperplasia or endometrial cancer, demonstrating a favorable safety profile.[21]

AIs decrease serum estradiol levels to less than 5 pmol/L (<2 pg/mL).[22] There are no RCTs demonstrating safety of HT in patients taking AIs, but one small study of six patients taking AIs by Kendall et al[23] showed that application of vaginal estrogen via low-dose (25 μg) Vagifem (Novo Nordisk A/S) tablets can increase these levels temporarily (up to 72 pmol/L) before decreasing back to less than 35 pmol/L (<10 pg/mL) after 4 weeks. Of note, the standard dose of Vagifem (Novo Nordisk A/S) is now 10 μg per tablet so these increases in estradiol levels are likely much lower. Regardless, it is unclear whether any small increase impacts the effectiveness of AIs.

A study from the United Kingdom showed that breast cancer survivors on AIs or tamoxifen using any formulation of low-dose vaginal estrogen did not have an increased risk of breast cancer recurrence after 3.5 years of follow-up.[24] Three other studies in patients using tamoxifen and four prospective studies in patients taking an AI have demonstrated low plasma estradiol levels in patients using various low-dose vaginal estrogen formulations, suggesting its safety in these patients. Given that estrogen levels in some studies were, however, significantly higher postintervention compared to preintervention, some authors still recommend using caution. In addition, most studies included small sample sizes and did not study long-term outcomes.[23–30]

Regarding SERMs, there may be a theoretical role for bazedoxifene (given in combination with CEE to form a tissue selective estrogen complex [TSEC]) in breast cancer survivors, as bazedoxifene plus CEE has been shown to be very effective for vasomotor symptoms and genitourinary syndrome of menopause (GSM), and bazedoxifene protects both the breast and endometrium.[31,32] Animal studies have shown that baxedoxifene inhibits growth of estrogen-dependent breast cancer cells, including cells that have developed resistance to tamoxifen, and actually targets the estrogen receptor for degradation.[33] However, there is no long-term data demonstrating safety in these breast cancer survivors. Ospemifene, a SERM originally developed for the treatment of osteoporosis, is an alternative to hormonal treatment for GSM. An RCT of 826 postmenopausal women found that oral ospemifene use for 12 weeks was more effective than placebo for symptoms of vulvovaginal atrophy,[34] and long-term follow up did not observe any breast or endometrial cancers.[35] More trials are needed to confirm its safety, and it is not currently approved by the FDA for women with breast cancer.

A 2016 Cochrane review showed that tibolone was more effective than placebo at treating vasomotor symptoms. There was no increased risk of breast cancer in patients with no personal history, but there was an increased risk of recurrence in breast cancer survivors.[36] Therefore, tibolone may be an alternative option in women without a history of breast cancer, but it is contraindicated in breast cancer survivors.

The North American Menopause Society allows that low-dose vaginal estrogen for treatment of GSM can be considered in patients with breast cancer who have failed nonhormone treatments, after consultation with an oncologist.[17] There is, however, no long-term data demonstrating safety in breast cancer survivors.

Breast Cancer and Progestins

Progestin type may affect breast cancer risk. A systematic review of breast cancer risk with use of micronized progesterone (MP) concluded that combined EPT with MP did not appear to increase breast cancer risk for up to 5 years.[37] One of the meta-analyses included found an RR of 0.67 (95% CI 0.55–0.81) with use of MP compared to synthetic progestins.[38] A second meta-analysis found that breast cancer risk was decreased with MP or dydrogesterone, but increased with MPA, norethisterone, or levonorgestrel.[39] The E3N study showed an increased risk of breast cancer with combined EPT regimens (RR 1.3, 95% CI 1.1–1.5). There was, however, no increased risk when the progestogen was oral MP (RR 0.9, 95% CI 0.7–1.2), and the increased risk appeared related to synthetic progestogen use (RR 1.4, 95% CI 1.2–1.7).[40] Several of the studies in the review found that breast cancer risk increased with MP use after more than 5 years of use, and while the risk with synthetic progestins was also increased after that time, the increased risk persisted after stopping synthetic progestins whereas the risk returned to baseline after stopping MP.[37]

Breast cancer risk may be mitigated by sequential progestin use as opposed to continuous use. In one prospective Swedish cohort study, continuous progestin use (both progesterone and testosterone derived) was associated with a two-fold increased risk of breast cancer, whereas there was no association with sequential use.[41]

Progestins appear to induce angiogenic growth factors, such as vascular endothelial growth factor, that cause neovascularization and cell proliferation in a tumor and can enhance the metastatic potential of breast cancer. Synthetic progestin exposure also leads to enrichment of cancer stem cells, which are highly tumorigenic cell populations in breast tumors that have stem cell–like properties.[42] Interestingly, Horwitz found that these stem cell–like tumor cells tended to be ER−/PR− but produced large colonies of tumor cells that acquired the receptors and were therefore ER+/PR+. Treatment of these colonies of cells with progesterone or MPA significantly increased the number of stem cell–like cells (from 1–3 cells per colony to 20%-30% and up to 100% in younger colonies). The authors hypothesized that small, young tumors may be most susceptible to stem cell activation by progestins. When activated, these stem cells (ER−/PR−) produce many more progenitor cells (ER+/PR+) which rapidly divide and lead to the growth of occult, noninvasive breast cancers that will further grow when exposed to estrogen, which has known proliferative properties. This theory therefore proposes that EPT reactivates cancer cells rather than directly causing cancer, and is consistent with findings that most of the breast cancers found after taking EPT are ER+/PR+.[16]

The increased risk of breast cancer associated with HT is likely related to the progestin component of EPT regimens. The type of progestin may be important, with MP possibly conferring much less risk, whereas synthetic progestins appear to significantly increase breast cancer risk. Some evidence also suggests that cyclic regimens may be safer than continuous regimens, although more studies are needed.