Menopausal hormone-replacement therapy and breast cancer risk: An updated and simplified view

INTRODUCTION

Menopause is a natural phenomenon wherein the menstrual cycle ceases at around 50 years in Western countries and 46 years in India,^[1-4] in response to the depletion of ova and fluctuation in hormone levels, i.e., lowered production of the E₂ and P₄, and elevated levels of pituitary follicle-stimulating hormone (FSH) and luteinizing hormone (LH) hormones.^[5-7] Such fluctuating hormonal milieu is responsible for various adverse symptoms and associated disorders that curtail the quality of life in menopausal women.^[6]

Menopausal hormone therapy (MHT), introduced in the 1960s, was extremely effective in managing menopausal symptoms. In the 1970s, a link between E₂ therapy and increased incidence of endometrial cancer in women with an intact uterus was found and resolved with combination MHT, i.e., a combination of E₂ and progestogen (generic term for progesterone and progestin). In the 1980s, the federal drug administration, USA approved MHT usage for managing menopausal symptoms and associated conditions.^[8]

MHT usage declined dramatically when the Women’s Health Initiative (WHI) published its findings in 2002, reporting excess risk of breast cancer (BC) with MHT, particularly in the combination regimen. A re-evaluation of the WHI trial data failed to find any statistical significance in the relative risk (RR) after adjusting for confounding factors.^[9] Ultimately, findings contradicting the WHI trials were published.^[10-12] Recent cohort studies, case-control studies, and meta-analyses have re-established the link between combined MHT usage and BC risk and also highlighted the risk associated with the type and timing of MHT usage.^[13-15] The relationship between MHT and BC risk in menopausal women reviewed herein is preceded by a brief discussion on menopause and its symptoms. We intend to convey this important topic to non-expert readers, especially the students.

THE MENSTRUAL CYCLE, MENOPAUSE, AND ITS ORIGIN

A normal menstrual cycle repeats every 28 days, beginning with the onset of menstrual discharge and terminating at the start of the next menses. It comprises three phases: Follicular (pre-ovulatory), ovulatory (release of oocyte), and luteal (post-ovulatory), driven by E₂, P₄, FSH, and LH.^[1,2]

The follicular phase is marked by the start of menstrual bleeding wherein, there is an overall decline of E₂ and P₄ levels and a slight increase in the pituitary FSH, stimulating the development of several ovarian follicles. Further, in the follicular phase, the dominant follicle matures and accumulates E₂-producing granulosa cells; rising E₂ level suppresses pituitary gonadotropin-releasing hormone secretion, lowering FSH.^[16,17]

The end of the follicular phase and initiation of the ovulatory phase are marked by surges of LH and FSH. The surge of FSH coinciding with the LH surge is lesser and function-wise not well understood.^[17]

In the luteal phase, in absence of fertilization, there is a decline in the level of LH and FSH levels, the ruptured follicle closes and forms corpus luteum, producing P₄.^[17]

Absence of fertilization results in the degeneration of the corpus luteum and termination of P₄ secretion. With the decline in E₂ level, endometrial breakdown, shedding and menstrual bleeding commence. During the late luteal phase, follicles for the next ovarian cycle are recruited, they resume growth and start secreting E_2, assisting the regeneration of the endometrial wall.^[2]

Figure 1 represents hormonal profile, ovarian and uterine changes, and important events of the menstrual cycle.^[17]

Close

Figure 1:

Hormonal profile, ovarian and uterine changes across menstrual cycle.^[17]

Export to PPT

By definition, menopause is an irreversible cessation of menstruation due to depletion of the follicular function of oocytes, indicating the end of natural female reproductive life. Stages of Reproductive Aging Workshop classifies ten reproductive stages. Five stages (−5–−1) before the FMP: final menstrual period, that is, Stage 0 followed by two more stages (+1–+2). −5–−3 is classified as the reproductive phase, followed by menopause transition, (−2–−1) characterized by amenorrhea, increased anovulatory cycles, variable cycle lengths, irregular FSH concentrations, and +1 and +2 as post-menopause.^[5] Although the size and number of follicles shrink, not all follicles become unresponsive. However, due to high FSH concentration, folliculogenesis is dysregulated. The decrease in the follicular number and size beyond the threshold causes anovulatory cycles lacking folliculogenesis followed by FMP; amenorrhea for 12 months.^[6]

According to the Indian Menopause Society, the different phases of menopause are natural menopause, premenopause, and perimenopause (indicative of the period exactly before and after 1 year of the final menopausal period (FMP).^[4]

Many theories and hypotheses have been put forward explaining the origin of menopause. Some theories suggest menopause to have originated during the Homo erectus period and the rest suggest it as a recent phenomenon.^[18] Broadly, these theories can be classified into two. One views it as an adaptive phenomenon or grandmother effect, according to which decreasing reproductive life span improves fitness.^[19] The other views menopause as an antagonistic pleiotropic phenomenon.^[19] According to the latter, during childbirth, ovarian hormones reach the highest level and drop within 24 h following parturition, generating a hypoestrogenic environment similar to menopause. Although both phenomena experience the same hormonal and epigenetic changes, these changes are beneficial during lactation and are detrimental during menopause. After childbirth, a low level of E₂ results in an elevation of serum lipoprotein concentration, which improves fat content in breast milk. In menopausal women, the same scenario increases the risk of cardiovascular diseases (CV), vascular inflammation, and arteriosclerosis. Dyslipidemia in menopausal women when combined with heightened serum inflammatory markers and hypertension affects mood and memory.^[19] In 2010, around 3.7 million cases of dementia were reported in India, of which 2.1 million were women. It is expected that such cases would be double in 2030. It has been shown that the risk of depression increases 2.5 times more in menopausal women rather than in premenopausal.^[4] After parturition, elemental calcium is displaced from the lumbar spine for the production of breast milk, which in menopausal women enhances the risk of osteoporosis.^[19] According to the Indian Menopausal Society, osteoporosis occurs in Indian women 10–20 years earlier than in Caucasians.^[4] Heat dissipation helps in heating the infant postpartum, which during menopause leads to hot flushes and other vasomotor discomforts.^[19] There is yet another theory that explains menopause as a consequence of reduced fertility associated with the accumulation of deleterious mutations in older women due to mating preference for younger females.^[19]

HORMONE REPLACEMENT THERAPY FOR MENOPAUSAL SYMPTOM MANAGEMENT

The psychological and physiological symptoms that accompany menopause, adversely affecting the quality of life, are listed in Table 1.^[20,21]

A longitudinal study by Penn Ovarian Ageing observed that 75% of postmenopausal women experienced joint pain and 73% suffered hot flushes.^[21] A multi-ethnic study assessed the bone mineral density and found it to decline rapidly during the transition and post-menopause phase; the rate being slower in African-American women compared to Asian women.^[22] The Vaginal Health: Insights, Views, and Attitude studied a large cohort of women and observed that more than 70% of menopausal women experienced some kind of vaginal discomfort; 24% experienced pain during intercourse, 35% observed involuntary urination, 22% suffered soreness, 15% experienced itching, and 17% suffered burning sensation.^[23]

During the normal reproductive cycle, E₂ levels fluctuate between 220 pmol/l and 918 pmol/l in the follicular phase and between 275 pmol/l and 1650 pmol/l in the luteal phase. During menopause, these fall below 41 pmol/l. Although MHT does not mirror an E₂ environment as that of the normal reproductive cycle, it produces enough for managing the climacteric symptoms.^[24] MHT successfully alleviates the symptoms that terribly affect the quality of life and also prevents diseases associated with E₂ deprivation. MHT can be categorized as E₂-only therapy (ERT) or combined E₂ + P₄ therapy (EPT). Table 2 lists the most common E₂, and P₄ preparations of MHT.^[24-26] ERT recommended for women without a uterus in formulations such as conjugated equine estrogen (CEE) and estradiol (E₂) is administered in many ways, orally, transdermally, intramuscularly, intranasally, subcutaneously, and locally. EPT, on the other hand, is prescribed for women with an intact uterus, either continuously or cyclically. This is because E₂ is mitogenic for endometrial cells and hence is combined with P₄.^[24-26]

Below, we will briefly discuss some of the menopausal symptoms and associated diseases that are successfully managed by MHT.

ASSOCIATION BETWEEN HORMONE REPLACEMENT THERAPY AND BC RISK

BC is one of the most common cancers in women. Globally, around 1.5 million, approximately 25% of women are diagnosed with BC every year. In India, it is ranked number one cancer among females. There are several risk factors, such as aging, family history, mutations in genes encoding breast cancer 1 and 2 gene (BRCA1/2), ataxia talangiectasia and Rad3 related (ATM), tumor protein p53 (TP53), and phosphatase and tension homolog deleted on chromosome 10 (PTEN), and high levels of growth factor receptors and steroid hormones (HER2 and estrogen, respectively).^[38] Another important risk factor for BC, especially in the postmenopausal setting, is metabolic syndrome (MetS), consisting of a combination of cardiovascular risk factors such as central obesity, hyperglycemia, hyperinsulinemia, hypertension, and hypertriglyceridemia. This is partly due to the association between insulin resistance and heightened estrogen/progesterone levels. Chronic oxidative stress and inflammation associated with MetS could also contribute to breast carcinogenesis.^[39]

Continued MHT usage is connected to a greater risk of BC. In addition, ERT increases the risk of endometrial cancer; hence, it is only limited to women who have undergone a hysterectomy. Women with an intact uterus are prescribed combined EPT.

In a trial conducted by the Women’s Health Initiative (WHI) in 2002, women with an intact uterus (n=8506) received a combined regimen of CEE and medroxyprogesterone acetate (MPA) and the hysterectomized settings (n = 5310) received CEE only. During the interventional phase, it was found that HR (hazard ratio) with 95% CI for BC was 1.24 in the combined regimen and 0.79 in the unopposed regimen. In the post-interventional phase, the HR for EPT was 1.32 and for ERT was 0.80. After a 13-year follow-up, the HR for EPT was 1.28 and that of ERT was 0.79, indicating a higher risk of BC in the combined regimen. Observation of a 26% increase in BC in women receiving EPT with a relative risk (RR) of 1.26 by WHI^[40] led to widespread panic and the downfall of MHT usage.^[11,41] However, in the WHI study, the RR was calculated from 30 cases/10,000 non-users versus 38/10,000 MHT users, representing a very minute absolute risk of 8/10,000 or 0.08% per year. Furthermore, the mean age of women in WHI was 63 years, but MHT is normally prescribed to symptomatic women below 60 years of age and around 70% of women in the study started MHT after entering the study. A re-evaluation of WHI results suggested that the risk is statistically insignificant after adjusting for confounding factors.^[9,42] Santen et al. developed biological and computer-aided models and calculated the percentage of de novo and occult tumors in the WHI study. He also analyzed the effects of EPT for 5 years or less on BC risk. Only 6% of the tumors diagnosed over a 5–7 year period occurred de novo, leaving 94% of pre-existing tumors growing in size reaching the diagnostic limit. This study concluded that HR for EPT used in the WHI trial was due to the pro-proliferative ability of the combination hormones to enhance the growth of already existing tumors rather than initiating new tumors. It was also found in this study that natural P₄ rather than synthetic progestogens is a safer choice.^[10,12]

Although the WHI study was criticized, recently several studies have confirmed a similar trend between MHT use and BC risk. It has also been found that prolonged exposure to either kind of MHT (ERT and EPT) results in a significant risk of developing BC. Kerlikowske et al. have observed a RR of 1.49 with 95% CI for BC risk in prolonged (more than 5 years) EPT users.^[43] BC risk concerning the type of MHT in the E3N-EPIC cohort observed in around 900 cases of invasive breast carcinoma out of 54,500 postmenopausal women during follow-up has been RR of 1.1 for ERT and 1.3 for EPT. It has also indicated that micronized progestogens are safer than synthetic progestogens as the RR for EPT (synthetic) has been 1.4 and the RR for EPT (micronized) has been 0.9.^[44] A meta-analysis of data from four randomized trials has reported 0.79 RR for ERT and 1.24 for EPT, indicating a higher risk of developing BC in EPT. This particular study has also suggested that the risk is higher in the present users than in past users.^[45] Another France-based large case-control study (1555 menopausal women, 739 cases and 816 controls) has confirmed the risk of BC concerning the type of MHT administered. In this study, the odds ratio (OR) for long-term (>4 years) ERT present user is 1.01, and 1.55 for EPT, suggesting that EPT enhances BC risk. It has also compared the OR for EPT with natural progestogen with the synthetic version and found a difference in values of 0.79 and 2.07, respectively. The study has further attempted to differentiate the OR of EPT with synthetic progestogen as progesterone-derived and testosterone-derived and reported these as 1.92 and 9.47, respectively. Finally, this study has revealed that continuous administration (OR = 2.7) of MHT poses a greater risk than cyclic regimen and the risk of developing ER+ and lobular tumor (OR=1.7 vs. 2.48) is high in EPT than in ERT.^[46] Another meta-analysis pooled information from 58 studies has analyzed the type and timing of MHT and BC risk. Among 143,887 postmenopausal women diagnosed with invasive BC, 51% were users of MHT, implying an overall risk of BC in MHT users compared to non-users. In this study, RR for ERT has been found as 1.33, with no significant difference among different estrogen supplements (CEE and E₂), or route of application, although, vaginal estrogen seemed to be safer a choice. For EPT, RR has been 2.08, almost similar for all forms of progestogens including micronized progestogens except dydrogesterone (RR = 1.39). Continuous EPT (RR = 2.3) has been found to be more harmful than cyclic (RR = 1.93), with the probability of ER+ tumor (RR [ERT] = 1.45, RR [EPT] = 2.4) being higher than the ER- type (RR [ERT] = 1.25, RR [EPT] = 1.42). Furthermore, risk of developing lobular tumor (RR [ERT] = 1.5, RR [EPT] = 2.7) has been more than ductal tumor (RR [ERT] = 1.25, RR [EPT] = 1.8). This study has also found that lean MHT users (RR [ERT] = 1.5, RR [EPT] = 2.36) are at higher risk of developing BC than obese women (RR [ERT] = 1.1, RR [EPT] = 1.6), even though higher body mass index (BMI) increases the risk of BC.^[13] A recent nested case-control study has observed an OR for long-term ERT to be 1.11 and 1.15 for CEE and E₂, respectively, and OR for long-term EPT to be 1.88, 1.79, 1.87, and 1.24 for norethisterone acetate, levonorgestral, medroxyprogesterone, and dydrogesterone, respectively. This study has reported a low risk associated with dydrogesterone, which further diminishes within 2 years of discontinuation, but this trend does not apply to other progestogens. It confirms a lower risk of BC with greater BMI and a linear increase in the chance of developing BC with a longer duration of use, as reported by other groups.^[14] The risk of BC recurrence in MHT users is unclear. A randomized trial Hormone Replacement Therapy after BC – is it safe? with a mean follow-up of 4 years suggests that HR for BC recurrence is 2.5 (95% CI) whereas the Stockholm trial suggests otherwise (HR = 0.8, 95% CI).^[47] Table 4 lists the studies reporting increased BC risk with MHT usage.

MOLECULAR AND GENETIC EFFECTS OF HORMONE REPLACEMENT THERAPY ON BREAST TISSUE

Progestogens administered during EPT such as MPA have been found to enhance the proliferation and metastasis of BC cells through genomic and non-genomic mechanisms. For example, it activates various signaling pathways such as mitogen-activated protein kinase or binds to steroid receptors, enters the nucleus, and activates or suppresses target gene expression, causing BC initiation or progression.^[48,49] With menopause, the ovary stops producing E₂, unlike adipocytes. The aromatase enzyme catalyzes the conversion of androgens to E₁ and E₂, while the steroid sulfatase and estrogen sulfotransferase convert E₁S to E₁ and E1 to E₁S, respectively. The 17-hydroxysteroid dehydrogenase enzyme is of two types, with Type I catalyzing the reduction of E₁S to E₂, whereas Type II catalyzing the reverse reaction.^[50] It was found that progestogens stimulate mRNA levels of E₂ -producing enzymes, elevating the BC risk.^[48] Horwitz and Sartorius suggested that the progestogens component of EPT reactivated the stem cell population in BC.^[51]

A Swedish study with 131 ER + BC samples observed alterations in 276 genes associated with MHT use, most of which were either involved in DNA repair or cell cycle regulation.^[52] Another study observed that BC risk in combined MHT users rose high in those women having one rare allele of CYP1A1 and CYP1B1 polymorphisms. The enzymes encoded by these two CYP450 are involved in the synthesis of catechol estrogen and estrogen quinone forming DNA adducts and causing enhanced proliferation of BC cells.^[53] A study based on data from the Normal Breast Study analyzed 54 MHT users (26 EPT and 12 ERT) and 36 non-MHT users. It identified 527 CpG sites (cytosine followed by guanine residues) located within 403 genes and 97 intergenic regions. It was found that DNA methylation was more prevalent in MHT users than non-users, which was further affected by factors such as duration and recentness of MHT usage. Strong modifications were observed in three CpG sites present on chromosomes 1, 2, and 3 in MHT users for non-users, two among these sites were hypermethylated and the other hypomethylated. Another CpG site (INPP4BCpG) was found to be highly methylated. Interestingly, loss of function of tumor suppressor gene INPP4B is predominant in many BC. These epigenetic modifications triggered by MHT might contribute to BC development and progression.^[54] Table 5 summarizes the molecular effectors of MHT contributing to increased BC risk.

CONCLUSION

Although MHT alleviates the majority of menopausal symptoms and improves the quality of life, the risk of developing BC is highly significant with its usage. The risk is greater with EPT compared to ERT and non-users, and it persists even after a decade of discontinuation. Factors such as genetic susceptibility, BMI, lifestyle, medical history type of regimen, dosage, duration, and individual response to MHT influence the associated BC risk. Thus, all risk factors must be considered and an appropriate MHT treatment strategy should be employed.