The cases detailed above highlight symptom burden faced by patients rendered surgically menopausal in the setting of gynecological cancer and management considerations that treating healthcare providers must consider when helping to optimize the quality of life for this population. Potential options for therapy, special considerations, side effects, and treatment regimens for vasomotor symptom management in gynecological cancer survivors will be discussed.

Approximately 25 % of gynecologic cancers occur in reproductive age women [1], many of whom will undergo an abrupt transition into menopause as a result of surgery, radiation therapy, or chemotherapy. While the ovaries in premenopausal women undergoing surgery for cervical cancer can often be spared [2], the surgical management of ovarian and endometrial cancers most often includes a bilateral salpingo-oophorectomy (BSO). Women undergoing pelvic radiation as primary treatment for cervical cancer will also lose ovarian function unless the ovaries are surgically transposed out of the radiation field prior to treatment [3]. An abrupt loss of ovarian function in young women is often associated with severe vasomotor symptoms (VMS), adversely impacting the quality of sleep and compounding their distress. Partially effective nonhormonal therapies can help and will be discussed elsewhere in this book. This chapter will concern itself with the role of estrogen therapy in the treatment of VMS in gynecological cancer survivors.

There are no definitive clinical data proving that estrogen therapy (ET) in patients with a history of uterine or ovarian cancer increases the risk or rapidity of tumor recurrence. There are, however, theoretical reasons for concern, including the presence of estrogen receptors in many gynecologic tumors, particularly in endometrial, ovarian, and uterine mesenchymal malignancies, as detailed below. In counseling those who might seek HRT to ameliorate VMS, providers may find useful guidance in concepts that emerged from the use of adjuvant antiestrogen therapy after surgery for breast cancer. It is clear that such therapy in women with estrogen receptor-positive breast cancer does improve disease-free survival when measured at an arbitrarily chosen point in time [4]. What remains uncertain is whether this clinical benefit of antiestrogen therapy is a function of enhanced death of residual microscopic disease or just a delay in growth and ultimate clinical recurrence. By extension of this argument to survivors of hormone receptor positive gynecological cancers, it also remains unsettled whether withholding hormone therapy will prevent a recurrence, or just delay its onset if one is destined to occur. At this point in time, there is no firm clinical evidence that hormone replacement therapy can cause a recurrence of a gynecologic cancer that would not have otherwise occurred, given adequate survival. There remains a clinical concern, however, that HRT could stimulate an earlier recurrence in survivors of hormone receptor-positive gynecological cancers, and this chapter will examine the evidence pertaining to specific gynecologic cancers.

Another important consideration for gynecologic cancer survivors contemplating hormone replacement therapy is the increased risk of venous thromboembolism (VTE) associated with such therapy and more specifically with standard estrogen dose formulations. This is particularly relevant to those whose baseline VTE risk is already elevated due to residual or active cancer, recent surgery, or ongoing adjuvant chemotherapy [5]. The potential additional risk of VTE should be assessed carefully with particular attention to dose and route of therapy in gynecologic cancer patients for whom estrogen replacement is felt to be necessary.

Endometrial cancer is well known to be a hormone-related malignancy, the risk of which is increased with long-term exposure to high levels of unopposed estrogen, be it from exogenous sources or endogenous ones that commonly prevail in obese women and those with chronic anovulation [6].

According to the Surveillance, Epidemiology, and End Results program of the National Cancer Institute (SEER), nearly 15 % of endometrial cancers occur in women under 50 years of age [7], and this percentage is likely to rise as obesity rates in young women continue to climb. Given the role of estrogen in the genesis of most endometrial cancers, there is a natural and justifiable concern about hormone replacement in endometrial cancer patients. This concern is supported by the fact that endometrial cancers express estrogen receptors, with a frequency that is related to histologic grade and type. In one immunohistochemical study, estrogen receptor (ER) positivity was localized within the malignant epithelium in 92 % of grade 1 endometrial cancers, 62 % of grade 2 tumors, and 12 % of grade 3 tumors [8]. In another study, well-differentiated endometrial adenocarcinomas expressed ER and progesterone receptors (PR) with higher frequency, and in much higher density, than poorly differentiated tumors [9]. Even in endometrial cancers not considered hormonally related, such as papillary serous tumors, ER was expressed in 30 % of cases [10]. In a randomized, placebo-controlled trial examining ET after surgery for endometrial cancer, 1,236 patients with stage 1 and 2 tumors were randomized to ET versus placebo [11]. The trial was closed prematurely after results from the Women’s Health Initiative (WHI) hormone trials were released making it difficult to accrue new patients and to maintain those in the treatment arm on estrogen [12]. Upon data analysis, the median age of subjects was 57 years, and the median follow-up duration was 3 years. Ninety-one percent of study subjects had grade 1 or 2 tumors, 15 % of those on ET had discontinued it within 6 months of enrollment, and only 41 % of patients randomized to the ET arm remained compliant with their assigned therapy for the entire treatment period. The cancer recurrence rates were 2.3 % in the ET arm and 1.9 % in the placebo arm, which was not a statistically significant difference. In addition, a number of smaller retrospective studies [13–16] have also found no worse outcomes from endometrial cancer in those treated with ET compared to those who were not, although conclusions are limited by selection bias inherent in retrospective study designs. In aggregate therefore, the existing clinical evidence provides no conclusive evidence regarding the absolute safety of ET use in endometrial cancer patients after surgery. However, the evidence does lend support to the position that symptomatic endometrial cancer patients whose baseline recurrence risk after surgery is low may be reasonably offered ERT for symptom relief after careful consideration and discussion of the pros and cons.

Progestin therapy has been used, with some success, as a treatment for patients with recurrent endometrial cancer [17]. Progestins are also effective for treatment of menopausal VMS [18]. Whether progestin added to estrogen could enhance efficacy against VMS and reduce recurrence risk in symptomatic endometrial cancer patients is an intriguing though speculative idea at this time. Ayhan et al. [19] prescribed daily 0.625 mg conjugated equine estrogen plus 2.5 mg medroxyprogesterone acetate to 50 endometrial cancer patients who initiated therapy 4–8 weeks after surgery. Forty-eight patients took the medication for at least 24 months with no documented recurrences in any of them. These theoretical advantages of progestins in hysterectomized endometrial cancer patients should be balanced, however, by concerns about increased breast cancer and cardiovascular risks associated with combination estrogen-progestin use, as were evident in the WHI estrogen plus progestin trial [12]. In the absence of firm clinical data about superior recurrence outcomes with combination estrogen-progestin therapies, this author prefers ET alone in carefully selected and counseled patients whose probability of disease recurrence is low.

There is also a lack of data regarding vaginal estrogen use in patients with a history of endometrial cancer. A study in breast cancer patients compared 75 users of vaginal estrogen, without oral estrogen, to matched nonusers and found no increased risk of recurrence, and clinicians treating gynecologic cancer patients may take some reassurance from this study [20]. Rahn and colleagues in 2014 reported a systematic literature review on vaginal estrogen and noted that with the exception of high dose vaginal estrogen cream (2 g daily of conjugated equine estrogen), other formulations of vaginal estrogen therapy that were tested did not increase serum estrogen levels beyond the normal postmenopausal range [21]. Based on these data, and the lack of demonstrated elevated recurrence risk even with systemic estrogen in endometrial cancer patients, low-dose vaginal estrogen appears to be a reasonable treatment for endometrial cancer survivors with symptoms of vulvovaginal atrophy resulting from estrogen deficiency.

Symptomatic menopause in ovarian cancer patients is not uncommon, because ovarian cancer surgery often includes BSO, and close to 30 % of new ovarian cancer patients are less than 50 years old [22]. Although most patients undergoing chemotherapy for ovarian cancer will have already undergone BSO, an occasional patient may have preservation of one normal ovary, but needs postoperative chemotherapy due to final pathologic findings, and chemotherapy has toxic effects on the ovaries and can induce menopause, either temporary or permanent. While ovarian cancer is not thought of as a hormone-dependent cancer to the same degree that is endometrial cancer, there are several theoretical concerns about HRT use in ovarian cancer patients, which have originated from epidemiologic studies and laboratory investigations.

From an epidemiologic point of view, an association between postmenopausal HRT and ovarian cancer risk has been known for some time. In a recent meta-analysis of 52 epidemiologic studies, the relative risk (RR) of ovarian cancer among current and recent HRT users was significantly higher than past and never-users [23]. The RR was 1.37 (95 % CI 1.29–1.46; p < 0.0001) and the increased risk was similar for estrogen-only and estrogen-progestin users, with the highest risk in current users. When data were stratified by histologic types, an increased RR was noted among HRT users for serous (RR 1.53, 95 % CI 1.40–1.66; p < 0.0001) and endometrioid (1.42, CI 1.20–1.67; p < 0.0001) cancers, but not for mucinous and clear-cell tumors. This discrepancy is interesting and relevant given that serous and endometrioid tumors are the ovarian cancers most likely to express estrogen receptors as discussed below. While these epidemiologic findings do support a possible cause–effect relationship between HRT and ovarian cancer, caution is advised due to inherent study limitations, including selection bias in two of the large studies that dominated the meta-analysis [24, 25]. Nevertheless, the epidemiologic evidence does raise concerns about HRT potentially provoking an earlier clinical recurrence in patients with occult ovarian cancer. It should be noted, though, that hormonal mechanisms possibly involved in new ovarian tumor initiation are likely different from those that could promote recurrence of treated ovarian cancer.

The expression of hormone receptors in ovarian cancers has been another source of concern when HRT is being considered. The rate of ER positivity varies markedly depending on the histology of the tumor, with 64 % of serous, 56 % of endometrioid, and 17 % of mucinous ovarian tumors expressing ER according to one series [26]. In a particularly large multisite immunohistochemical and microarray study which included 2,933 patients with invasive epithelial ovarian cancer, strong expression of ER alpha and progesterone receptors was found in over 60 % of serous and endometrioid tumors, but only in about 15 % of mucinous and clear-cell tumors [27]. Other studies have corroborated these findings and found that ER expression in serous tumors is also related to degree of tumor differentiation (and thus inversely with tumor aggressiveness), with expression rates exceeding 90 % in borderline tumors, greater than 50 % in grade 1 tumors, and under 30 % in grade 2–3 tumors [28, 29]. Furthermore, tissue culture studies on ER-positive ovarian cancer cell lines demonstrated increased growth with an addition of 17-beta-estradiol, which was not observed in ER-negative cell lines and which was blocked with addition of tamoxifen, a selective estrogen receptor modulator [30]. Another tissue culture study demonstrated that the growth rate with estradiol was enhanced in ovarian cancer cell lines expressing ER alpha but not ER beta [31].

In the clinical setting, the use of antiestrogen therapies in women with recurrent ovarian cancer has been attempted, with variable results. One study reported a 17 % partial CA125 response rate (>50 % decrease in the serum level) and a 9 % partial radiologic response rate to an aromatase inhibitor (letrozole), with subgroup analysis showing higher response rate correlating with higher tumor expression levels of ER [32]. Another study however showed only a 2 % partial response rate to another aromatase inhibitor anastrozole, with a mixture of ER-positive and ER-negative tumors [33]. As to borderline serous ovarian tumors, there is paucity of data, but in one study, two women with relapsed chemotherapy-resistant disease were reported to achieve complete long-term remission with anastrozole therapy [34]. At present, it is appropriate to conclude that aromatase inhibitors have real potential value in the treatment of grade one ovarian cancers (specifically endometrioid and serous tumors) and serous borderline ovarian tumors, but further study is warranted.

While tumor response to antiestrogens does not prove that ET will of necessity accelerate growth of tumor in patients, it does lend some support to that concern. On the other hand, studies that have directly analyzed HRT use in ovarian cancer patients do provide some reassurance. In one retrospective analysis [35], 78 ovarian cancer patients treated postoperatively with HRT were compared with 295 patients who were not treated with hormones; ET alone was used in 32 patients, estrogen and progestin in 38 patients, and progestin or testosterone alone in 8. Cox regression analysis was done to address differences in prognostic factors. There was no statistically significant difference between the groups in terms of survival or disease-free survival. In a randomized trial of 130 patients with advanced stage, high-grade epithelial ovarian cancer [36], patients were randomized to oral conjugated estrogen (0.625 mg daily), or not, after surgery. Nineteen patients were lost to follow-up or stopped taking the estrogen or started estrogen in the group randomized to no estrogen. After more than 4 years of minimum follow-up, there was no adverse effect found in the estrogen group in terms of recurrence rate, disease-free survival, or overall survival. In 2015, the results of a randomized trial with a median follow-up of 19 years in surviving patients were reported [37]; 150 patients with any stage ovarian cancer were randomized to HRT or no HRT after surgery. The median age was 59 years. The hormone replacement was planned for a minimum of 5 years, if tolerated; 53 patients received estrogen alone, 19 patients received estrogen plus norgestrel. Forty-six out of 72 patients discontinued the HRT at some point during the follow-up period, and the median estimated time taking the HRT for patients assigned to this arm was 1.14 years. Sixty-seven percent of the patients in the HRT group died of ovarian cancer versus 75 % in the control group with the overall survival being significantly higher in the HRT group (hazard ratio, 0.63; 95 % CI, 0.44–0.90; p = .011). The authors concluded that women with severe menopausal symptoms after treatment for ovarian cancer can safely take HRT without compromising their survival.

Adult granulosa cell tumors (GCTs) are uncommon ovarian cancers, presenting in both premenopausal and postmenopausal women. They are indolent tumors which usually present at an early stage, and while most are cured with surgery, very late recurrences have been reported [38]. Between 32 and 66 % of GCTs are ER-positive [39], leading to concerns about stimulating tumor recurrence with ET in patients who have undergone BSO as part of treatment. The typically long latency period makes the issue of provoking an earlier recurrence more problematic, and reports of tumor response to antiestrogenic therapy [39] also support a degree of concern regarding potential pathophysiological relevance of estrogen for these tumors.