Endocrine consequences of breast cancer therapy and survivorship

Abstract

Breast cancer survivorship is increasing, due to earlier diagnosis of the disease and more effective therapies. Long-term endocrine sequelae, including early menopause, bone health, fertility implications and menopausal symptoms, are important survivorship issues. Ovarian failure is common with chemotherapy and options for preserving fertility in young women include ovarian suppression during chemotherapy and oocyte or embryo cryopreservation before chemotherapy. Tamoxifen as adjunct therapy in premenopausal women leads to ovarian stimulation, sometimes ovulation and occasionally pregnancy with important teratogenic implications. Aromatase inhibitor therapy with or without gonadotrophin releasing hormone (GnRH) agonist leads to profound bone loss and anti-resorptive therapy is advised to prevent fracture. Tamoxifen acts to preserve bone in postmenopausal women but not premenopausal women. Pregnancy is not discouraged in young women with early breast cancer, even to the point of pausing adjunct therapy in order to conceive. However, menopausal hormone therapy is discouraged even years later. Non-hormonal therapy for menopausal symptoms in breast cancer survivors is available but, in some cases, estrogen-containing therapy may be worthy of consideration for quality of life in the informed patient.

Keywords:

• Breast cancer
• survivorship
• tamoxifen
• aromatase inhibitors
• adjuvant therapy
• premature menopause
• fertility
• bone density

Introduction

Breast cancer is the most commonly diagnosed cancer worldwide [1], and the most commonly diagnosed malignancy in women 40 years of age and younger [1]. The incidence is increasing, mainly due to improved screening techniques, but survival rates in developed countries such as the UK, the USA and Australia are improving [1]. Breast cancer mortality had minimal change from the 1930s through to the 1970s when radical mastectomy was the primary mode of treatment. Survival rates improved in the 1990s when breast cancer screening programs commenced and effective medical therapies became available. Age-standardized breast cancer mortality dropped by 40% between the 1980s and 2020 in high-income countries [2], with current 5-year survival rates of 92% in Australia [3]. Widespread screening and more sensitive breast imaging techniques have increased the diagnosis of early breast cancer, and increased the detection of node-negative disease and of estrogen receptor-positive disease with a better prognosis, albeit at the cost of possibly detecting disease that would not ever become life-threatening [4]. Nevertheless, mortality rates of non screen-detected breast cancers have paralleled those seen with screen-detected breast cancers, suggesting it is not only improved screening that is responsible for increased survival [4]. Given the improved prognosis of patients with breast cancer, survivorship issues including menopausal symptoms, bone health and fertility implications are becoming more important. For some of these, there are limited acceptable and effective therapeutic options.

Many breast cancer treatments, especially for hormone-positive breast cancer, reduce estrogen production or effect. Such therapies include chemotherapy-induced ovarian failure or insufficiency, bilateral oophorectomy, gonadotrophin releasing hormone (GnRH) analogs, tamoxifen and aromatase inhibitors. Given the improved long-term survival from breast cancer therapy, it is incumbent upon clinicians to take note of the consequences of these therapies on long-term health [5].

Menopausal symptoms including hot flushes, mood changes, insomnia, vulvovaginal atrophy (VVA), sexual dysfunction and reduced quality of life are common in breast cancer survivors, due to treatment-induced early menopause, a reduction in estrogen levels or cessation of menopausal hormone therapy (MHT). Iatrogenic menopause may be associated with more severe symptoms than physiological menopause [6], probably reflecting the rapidity of the menopause transition and the magnitude of estrogen deficiency [6]. Typically, hormone therapy is deemed to be contraindicated in women with a history of hormone-responsive cancer including breast cancer, irrespective of the hormone receptor status of the tumor [7]. However, non-hormonal therapies are often less effective for treating climacteric symptoms, especially hot flushes and VVA, than estrogen [8].

In this review, we aim to summarize the pertinent endocrine sequelae in breast cancer survivors. We will also pose questions for future research which will be of increasing importance in this growing group of patients.

Chemotherapy-induced ovarian failure and infertility

Chemotherapeutic agents in premenopausal women are toxic to the ovary and may cause early menopause and infertility. Triple-agent regimens such as cyclophosphamide, methotrexate, fluorouracil (CMF) carry a high risk of infertility if administered for more than four cycles in women older than 40 years, with a lower risk in younger women [9]. Suggested mechanisms by which chemotherapy compromises follicular ovarian reserve are accelerated ovarian follicle maturation, direct quiescent follicle DNA damage and disrupted ovarian vascularization [9]. Given the high incidence of breast cancer in young women and the fact that breast cancer treatments commonly affect ovarian function, fertility preservation and subsequent childbearing are of paramount importance to many patients. This is especially so due to the later age of childbearing, meaning many women have not completed their families at the time of diagnosis of breast cancer.

GnRH analogs may be used to preserve ovarian function during chemotherapy but the outcomes on fertility preservation are inconsistent and long-term safety results are not available [10]. GnRH analogs have been shown to reduce the rate of premature ovarian insufficiency in premenopausal women with early breast cancer on adjuvant chemotherapy. A systematic review and meta-analysis included 873 patients from five trials, with the premature ovarian insufficiency rate found to be 14.1% in the GnRH analog group and 30.9% in the control group [11]. In the same publication, at least one post-treatment pregnancy was achieved in 10.3% of the GnRH analog group and in 5.5% of the control group. This suggests that temporary ovarian suppression with GnRH analogs during chemotherapy is a useful strategy to preserve ovarian function and fertility in premenopausal women with early breast cancer.

However, oocyte and embryo cryopreservation have become standard of care prior to initiation of systemic breast cancer therapy in women wishing future conception. This is usually timed after surgery and before adjuvant therapy. The delay in adjuvant therapy has not been shown to compromise survival [12]. In carriers of BRCA mutations it has the added advantage of pre-implantation testing of embryos to screen for carriers of those mutations [13]. However, in some cases, a delay in cancer treatment to pursue these interventions may be undesirable, and oocyte and embryo cryopreservation do not prevent the development of chemotherapy-induced premature ovarian insufficiency. Furthermore, not all women have access to and the resources for this assisted reproductive technology.

Safety of pregnancy in premenopausal women with breast cancer

In light of the complex issues already outlined relating to premenopausal women with breast cancer who have not completed their families, the POSITIVE (Pregnancy Outcome and Safety of Interrupting Therapy for Women with Endocrine Responsive Breast Cancer) trial was designed to assess the safety of a temporary interruption of endocrine therapy to attempt pregnancy. This international trial of women aged 42 years or younger with stage I, II or III hormone receptor-positive breast cancer who had been on adjuvant endocrine therapy for between 18 and 30 months evaluated outcomes in women who were able to interrupt endocrine therapy for up to 2 years to allow for pregnancy, conception, delivery and breastfeeding [14]. Among 497 women who were followed, 74.0% had at least one pregnancy. At 1638 patient-years of follow up, 44 patients had a breast cancer event, which was within the prespecified safety threshold of 46 events. The median follow-up was 3.4 years. Whilst this trial gives some reassurance regarding the safety of interrupting therapy in the short to medium term, it was not, and is not feasible to be, a randomized controlled study. The majority of women (85%) resumed adjuvant endocrine therapy after breast-feeding, and it is possible that healthier women were referred to or chose to participate in the study. Longer follow-up is clearly needed. Prior to the POSITIVE trial publication, previous retrospective data had shown no clear evidence of worse survival among women who became pregnant or had a live birth after breast cancer than those who did not [15]. Given the apparent safety of a high-estrogenic state in pregnancy in relation to breast cancer outcomes, can we then feel justified in questioning whether a complete contraindication to exogenous estrogen for later menopausal symptom management is warranted?

Adjuvant therapies causing a reduction in production or effect of estrogen

Estrogen deprivation therapy

GnRH analogs downregulate luteinizing hormone (LH) and follicle stimulating hormone (FSH), thereby switching off all gonadotrophin-dependent ovarian hormone production (both estrogens and androgens) in a reversible fashion. Bilateral oophorectomy, similarly, induces the rapid onset of permanent iatrogenic menopause with both estrogen and ovarian androgen deprivation.

Aromatase inhibitors in postmenopausal women cause near-complete deprivation of circulating estradiol by blocking the conversion of ovarian and adrenal androgen to estrogen [16]. In premenopausal women, aromatase inhibitors are only effective when used in combination with ovarian function suppression either via oophorectomy or GnRH analogs [17]. Without ovarian suppression they act to induce unopposed estrogen production from the premenopausal ovary, as in seen with their use in ovulation induction [18].

Estrogen antagonism therapy

Conversely, tamoxifen – a selective estrogen receptor modulator – antagonistically competes with estrogen for binding sites in breast tissue and causes anti-estrogenic and antitumor effects [19]. It is the preferred adjuvant therapy in premenopausal women, with or without GnRH agonist ovarian suppression [20].

Tamoxifen carries some unique ovarian, fertility and pregnancy considerations. Tamoxifen acts as an estrogen agonist in the uterus and induces growth of endometrial cells, as an agonist in the liver to raise high-density lipoprotein (HDL) cholesterol and lower LDL cholesterol, and as an agonist in inducing coagulation factors [21]. However, tamoxifen acts as an antagonist in the hypothalamus and pituitary with actions similar to clomiphene leading to ovarian hyperstimulation in premenopausal women, high serum levels of 17β-estradiol and increased incidence of ovarian cysts [22–24]. This raises the question of whether the very high serum estradiol concentrations with ovarian hyperstimulation in this setting are reliably blocked by the competitive antagonistic action of tamoxifen at the breast. The relationship between high estrogen levels in such cases and risk of relapse of breast cancer is an important issue worthy of further research.

Similar to clomiphene, tamoxifen can induce ovulation [25] and has been associated with on-treatment pregnancy. Appropriate contraceptive options are limited but highly important in premenopausal women with breast cancer, especially those who are on tamoxifen without ovarian suppression. The CANTO study of women with stage I–III breast cancer, including 2900 premenopausal women, showed that approximately 90% of patients chose to use reversible mechanical methods of contraception in the 2 years after diagnosis of breast cancer, with male condoms being the most commonly used [26]. Despite the relatively low effectiveness of condoms, they are perhaps the best available option with hormone-containing methods like the oral contraceptive pill and progestogenic compounds being contraindicated.

If a woman falls pregnant while taking tamoxifen, the potential risks require careful counseling. The general advice is to stop tamoxifen treatment due to its reported teratogenic effects [27]. Women need to weigh up the options of coming off their cancer treatment, which may increase risk of breast cancer recurrence, versus their desire for a child, and the risks to the fetus of tamoxifen exposure in utero. Animal studies have shown various fetal toxicities from tamoxifen during pregnancy, similar to some of the genital tract anomalies seen with diethylstilbestrol exposure [28]. A literature review and database of the Netherlands Pharmacovigilance Centre Lareb and the database of the International Network on Cancer, Infertility and Pregnancy found a total of 238 cases of tamoxifen use in pregnancy [28]. Of the 167 pregnancies with a known outcome, 21 were complicated by an abnormal fetal development; that is, 12.6% compared to 3.9% of the general population. The authors described the malformations as non-specific although they highlighted that long-term follow-up will be required.

To further complicate decision-making around fertility options in young women, waiting until they have finished their course of endocrine treatment in 5–10 years would reduce their likelihood of successful pregnancy outcomes due to advancing maternal age.

Bone health considerations with endocrine therapy in breast cancer

Estrogen deprivation in breast cancer treatment is associated with accelerated bone loss and increased fracture risk. With aromatase inhibitor use in postmenopausal women, circulating estrogen levels are reduced by >98% [16]. The Anastrozole, Tamoxifen, Alone or in Combination (ATAC) trial found that in postmenopausal women with breast cancer, anastrozole treatment was associated with a significant improvement in disease-free survival but had higher fracture rates compared with tamoxifen. Fractures were more frequent during active treatment (odds ratio 1.33, 95% confidence interval 1.15–1.55, p < 0.0001) in women receiving anastrozole compared with tamoxifen, but were similar in the post-treatment follow-up period [29]. After 2 years of anastrozole treatment, there was a median 4.1% loss of bone mineral density (BMD) at the lumbar spine and 3.9% loss at the total hip, with increased bone remodeling indicated by increased N-telopeptide (NTx) and bone alkaline phosphatase (bALP) [30]. A meta-analysis of 21 randomized controlled trials and cohort studies published in 2018 of women aged 65 years and under with non-metastatic breast cancer found a 17% higher fracture risk in the aromatase inhibitor group than in the no aromatase inhibitor group [31].

In premenopausal women, since aromatase inhibitors induce ovulation they can only be used with concurrent ovarian function suppression. Combined use of ovarian function suppression and aromatase inhibition in this group results in the most profound suppression of estrogen production and the most pronounced bone loss among all types of endocrine therapies for breast cancer [32]. In a study of premenopausal women with endocrine-responsive breast cancer, women who received anastrozole and goserelin had a reduction in BMD of 11.3% at the lumbar spine and 7.3% at the trochanter after 3 years [33]. Two years after completing treatment, there was a partial improvement but recovery to baseline BMD was not achieved.

Conversely, tamoxifen has variable effects on skeletal health depending whether it is used in premenopausal or postmenopausal women. In premenopausal women, tamoxifen acts as a partial antagonist in bone and has a detrimental effect on bone remodeling by preventing the more potent activity of endogenous estrogen, whereas in postmenopausal women with low circulating estrogen concentrations, tamoxifen has a partial agonist effect and prevents bone loss [32]. Only modest increases in BMD have been seen with tamoxifen in postmenopausal women, and fracture outcome data have been inconclusive among trials. A systematic review and meta-analysis showed that fracture risk did not differ between the tamoxifen and no-tamoxifen groups (pooled relative risk 0.95, 95% confidence interval 0.84–1.07) [31]. In a trial of premenopausal women with early breast cancer receiving adjuvant endocrine therapy, there was a significant reduction in total body BMD (–1.5%) after 2 years of tamoxifen treatment [34], compared with a 5% reduction in BMD in a group receiving goserelin. When goserelin and tamoxifen were used together, the tamoxifen appeared to attenuate the GnRH analog effect (total body BMD −1.4%).

There are a number of guidelines on the management of bone health in women with breast cancer receiving endocrine therapy, proposing indications for initiation of anti-resorptive therapy as well as suggesting which anti-resorptive to use [16,35,36]. All women should be assessed for general osteoporosis risk factors, and postmenopausal women starting aromatase inhibitors and premenopausal women starting tamoxifen should have a baseline bone density. Standard lifestyle advice is usually given, including smoking cessation, weight-bearing exercise and adequate calcium intake and sufficiency of vitamin D. Although there is inconsistency regarding the dosing schedule and duration of anti-resorptive treatment, there is general agreement in the guidelines that a t-score of −2.0 is an indication for intervention. The rebound resorptive effect and increased fracture incidence with cessation of denosumab is not fully articulated in the guidelines and neither is, given its long half-life, the safety of bisphosphonate therapy for future pregnancies [37,38]. These factors are worthy of future research to better inform appropriate bone treatments in women with breast cancer.

Menopausal symptoms in breast cancer survivors

Management of menopausal symptoms in breast cancer survivors can be a significant challenge. The type of breast cancer treatment given and the age of the woman can be significant determinants of the severity of menopausal symptoms, which can be a major cause of reduced quality of life in women with breast cancer. With increased survival rates and many women having years of life ahead following diagnosis of breast cancer, ensuring quality of life factors are prioritized is of high importance. Women experiencing significantly reduced quality of life may be more likely to discontinue their breast cancer treatment, potentially increasing risks of recurrence and reduced survival [39].

Non-hormonal therapies

Whilst there are a number of non-hormonal treatment options for menopausal symptoms, it is well recognized that these are generally less effective than estrogen. The recent North American Menopause Society Position Statement list those considered to have efficacy in randomized studies as cognitive-behavioral therapy, clinical hypnosis, selective serotonin reuptake inhibitors/serotonin–norepinephrine reuptake inhibitors, gabapentin, oxybutynin, weight loss, stellate ganglion block and neurokinin 3 receptor blockade [40]. The last and most recent modality stems from the discovery of the role of kisspeptin, neurokinin B, and dynorphin (KnDY) neurones in the control of both reproduction and thermoregulation. The finding of the increased numbers and expression of these neurones in the hypothalamus of postmenopausal versus premenopausal women and the experimental administration and blockade of neurokinins in producing and eliminating vasomotor symptoms, respectively, have identified this as an area for drug development [41]. Recent reports of clinical trials of pharmacological agents which act on the KnDY pathways for the control of vasomotor symptoms have shown promise for effective non-hormonal management of vasomotor symptoms, although only the study by Vrselja et al. has been conducted in women with breast cancer on adjuvant therapy [42–45].

Estrogen-containing therapy

In women with a personal history of any type of breast cancer, systemic MHT has traditionally been contraindicated [46]. Is a blanket ban on MHT appropriate, or is there room for individualization?

A systematic review and meta-analysis of systemic MHT in breast cancer survivors evaluated four randomized controlled trials including 4050 patients, and found an increased risk of breast cancer recurrence in women taking estrogen/progestogen combination or tibolone compared with placebo (hazard ratio 1.46, 95% confidence interval 1.12–1.91, p = 0.006) [7]. However, a subgroup analysis showed an increased risk of breast cancer recurrence in women with hormone receptor-positive disease but not in those with hormone receptor-negative tumors. Closer examination reveals important differences between hormone regimens in the studies reviewed. Three studies using estrogen-containing formulations used medium-dose estrogen, either estradiol 2 mg or conjugated estrogens 0.625 mg, but with different progestogen exposure. Only the tibolone study and the HABITS study reaching statistical significance for breast cancer recurrence. Tibolone is, in essence, an androgenic progestogen with estrogenic, progestogenic and androgenic metabolites [47]. The HABITS study employed mostly continuous combined estrogen and norethisterone acetate [48]. However, the Stockholm Study and the study by Vassilopoulou-Sellin et al. limited or avoided progestogen exposure [49,50]. These data and the contrasting findings of the two Women’s Health Initiative (WHI) studies – estrogen alone versus estrogen plus progestogen – suggest that progestogen exposure is important not only in risk of breast cancer but also in breast cancer recurrence [51,52].

A recent publication by Bluming argued that only 1 of 25 studies, the HABITS trial, of estrogen-containing MHT after breast cancer diagnosis, published between 1980 and 2013, showed an increased risk of breast cancer recurrence in the active treatment group versus placebo and that was limited to local or contralateral and not distant recurrence [53]. None of the studies (including the HABITS study) reported increased breast cancer mortality. In this respect, Bluming stated that women with a history of breast cancer are being unfairly denied MHT despite its superior effects in controlling menopausal symptoms as well as its other benefits, including for bone health. An absolute ban on MHT for breast cancer survivors does seem incongruous given that pregnancy is not absolutely discouraged.

Whilst the aforementioned studies have used medium-dose estrogen and variable progestogen formulations and doses, the question remains whether low-dose estrogen and different progestogens would be safer. The emerging availability of estetrol, a weak natural estrogen with a specific profile of estrogen receptor-α activation, may present further therapeutic opportunities for use in women with breast cancer when more extensive clinical studies have been completed. Currently, the use of estetrol/drospirenone contraceptive is contraindicated in women with breast cancer or a history of breast cancer, although some data have shown short-term exposure to estetrol induced apoptosis rather than proliferation of breast carcinoma cells in women with recently diagnosed breast cancer [54]. This is a research question worthy of further investigation. Furthermore, the safety of the additional progestogen should be considered. In women who have had a hysterectomy, estrogen-only therapy is appropriate. For women with an intact uterus who require the addition of progestogen, the breast cancer risk with different progestogens has been reported in observational studies and reviews suggesting that natural micronized progesterone, rather than synthetic progestin, is safer in terms of breast cancer risk and, therefore, possibly breast cancer recurrence [55,56].

Locally applied vaginal estrogen is an effective treatment for VVA in women after menopause, with genitourinary symptoms often being particularly severe in women taking aromatase inhibitors. However, the use of vaginal estrogen in women with breast cancer has also been a controversial topic. Observational data suggest relative safety of vaginal estrogen in women with postmenopausal estrogen receptor-positive non-metastatic breast cancer [57], although a subgroup analysis showed an increased risk of recurrence but not mortality in women receiving vaginal estrogen with adjuvant aromatase inhibitors. The dose of vaginal estrogen was not clearly outlined, given this was an observational cohort study where women were classified as being users of vaginal estrogen therapy based on at least two redeemed prescriptions. However, it would be prudent to carefully consider the dose and type of estrogen used, with Stanczyk et al. outlining the limitation of existing studies examining the endometrial safety of vaginal estrogen, which have all had a short duration of follow-up [58]. They suggest that studies should be carried out which quantify endometrial and serum estradiol, estrone and equine estrogen concentrations by mass spectrometry after low-dose vaginal estrogen administration in different doses and formulations in order to further investigate the endometrial exposure, systemic exposure and, therefore, safety of this intervention. A randomized, double-blind, placebo-controlled trial of ultra-low-dose 0.005% estriol vaginal gel for the treatment of VVA in postmenopausal women with early breast cancer treated with non-steroidal aromatase inhibitors found an improvement in symptoms and signs of VVA in these women [59]. They found estriol levels increased initially and normalized by week 12, whilst estradiol and estrone remained mostly undetectable throughout the study, on measurement by liquid chromatography–tandem mass spectrometry. Notably, there are no US Food and Drug Administration (FDA)-approved formulations of estriol available. In a departure from using estrogen for VVA, Thurman et al. report the use of intravaginal tamoxifen, on the basis of its known selective estrogen agonistic effect on vaginal tissue [60]. They report a decrease in symptoms, a decrease in vaginal pH and a improvement in vaginal maturation index with intravaginal tamoxifen exposure.

Compounded bioidentical MHT is not recommended for use in postmenopausal women, and even less so in women with a history of breast cancer [61]. These products are not subject to quality control, or to the same tests of safety, efficacy and dosing consistency as regulated MHT. For estrogen, Newman et al. have shown that compounded formulations achieved lower systemic concentration compared with FDA-approved products risking inefficacy in the intended therapeutic endpoints [62]. Excipients, contamination and addition of extraneous compounds have all led to adverse events with compounded bioidentical hormone therapy [63].

With the current evidence available, some recommendations are allowing MHT to be considered in women with severe symptoms and poor quality of life, when non-hormonal treatments have failed and after a detailed discussion with the patient and her oncologist about the potential risks and benefits of treatment [64]. In particular, the North American Menopause Society Position Statement 2022 states that ‘if symptoms are severe and unresponsive to nonhormonal options, women, in consultation with their oncologists, may choose hormone therapy after being fully informed about the risks and benefits’ [65,p.781].

Conclusions

With increasing survivorship of women with breast cancer, long-term endocrine sequelae are becoming more common. Important considerations are menopause and its symptoms, bone health, and fertility and pregnancy in younger women. Issues related to tamoxifen, including ovarian hyperstimulation and risk of fetal malformations, present a conundrum in premenopausal women. The high estrogen levels that can occur with tamoxifen treatment warrant further evaluation as to whether this adversely effects cancer outcomes. The previous contraindication to MHT use in breast cancer survivors should be reconsidered, as there are some women in whom non-hormonal treatments are inadequate. Quality of life is of vital importance, and women should not be denied a treatment that can improve quality of life if they have been adequately informed of the risks. More research to guide the safe use of MHT in breast cancer survivors would be helpful for clinicians managing these women.

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability statement

Data referred to in this manuscript are derived from public domain resources as listed in the references.

Additional information

Funding

None.