Leon Speroff Professor and Vice Chair for Research, Department of Obstetrics and Gynecology, Oregon Health & Science University, Portland
Dr. Jensen reports he receives grant/research support from and is a consultant for ObstetRx, Bayer, Merck, and Sebela Pharma; is a consultant for AbbVie, Mithra, and Daré Bioscience; and receives grant/research support from CooperSurgical and Population Council.
To practice as a consultant in reproductive medicine, clinicians require a strong background in hormonal therapy. In this feature, we will review the role of steroidal estrogens used in contraceptive and hormone therapy.
First, let’s briefly review how steroid hormones work. Most steroid hormones circulate through the blood bound to specific carrier proteins. For example, estrogens and androgens circulate bound to sex hormone binding globulin (SHBG), while glucocorticoids and progesterone bind to cortisol binding globulin (CBG). Albumin binds all steroid hormones with low affinity but high capacity. Only the free (unbound) hormone can exert a biologic response.1 Route of administration, dose, metabolism, and tissue-specific receptor activity of various natural and synthetic steroid and peptide hormones also influence the biologic response. The lipid-like nature of steroid hormones allows them to pass through cell membranes easily.
The classical mechanism of steroid action requires hormone binding to a specific hormone receptor. Two nuclear estrogen receptors have been described: ERα and ERβ. Receptor binding leads to dimerization and binding with other regulatory cofactors, followed by interaction with nuclear DNA to stimulate transcription of mRNA. The mRNA is transported to the cytosol and ribosomes for translation to protein.2 Although there is some evidence that steroid hormones also work through direct (non-nuclear) actions, the take-home point is that classical steroid hormone action takes time (hours to days). In other words, the clinical response following administration of a steroid requires patience.
Estrogen is not a hormone. Rather, the term estrogen encompasses a family of natural and synthetic hormones with activity at estrogen receptors. Four natural estrogens exist in humans: estrone (E1), estradiol (E2), estriol (E3), and estetrol (E4). These 19-carbon steroids differ by the number of hydroxy (-OH) groups present on the cyclopentanophenanthrene ring steroid backbone, hence the shorthand nomenclature.
E2, the most biologically active of the natural estrogens and the primary estrogen of the reproductive years, is secreted by the ovaries from menarche through menopause. The hypothalamic-pituitary ovarian axis regulates estrogen production by stimulating ovarian follicle development. Theca cells surrounding antral follicles produce androstenedione and testosterone. Granulosa cells then aromatize these androgens to estrone and estradiol. Isomerization of estrone to estradiol occurs through 17b-hydroxysteroid dehydrogenase.
E1, the primary estrogen of menopause, results primarily from the conversion of adrenal androstenedione by aromatase in peripheral fat. E1 is about 12-fold less potent than E2.3 However, since E1 can undergo isomerization to E2, peripheral conversion in young and old obese women can result in physiologically and clinically important levels of E2.
E3 is the primary estrogen produced by the placenta. E3 is a weak estrogen (80-fold less than E2) that undergoes rapid elimination.3 Prior to the development of fetal monitoring, obstetricians monitored E3 levels to evaluate fetal well-being. Another pregnancy-related estrogen, E4, is the most recently discovered natural human estrogen. The fetal liver produces E4 beginning in the ninth week of pregnancy and production ceases during the first week after birth. E4 is about 30- to 35-fold less potent than E2.4 It makes sense that these lower-potency estrogens serve to reduce the effect of the high estrogen environment of pregnancy on the fetus.
When given orally, E1, E2, and E3 undergo rapid hepatic metabolism through conjugation for excretion. Estradiol has a half-life of about 14-16 hours, with estrone and estriol even shorter. The poor oral bioavailability and rapid metabolism of natural estrogens presents a challenge when using these agents for oral therapy. In contrast, E4 undergoes minimal liver metabolism and has a half-life of about 28 hours.4
Structural modifications of natural estrogens with polar sidechains designed to improve bioavailability with oral dosing have been synthesized. These can be considered prodrugs. For example, estradiol valerate (E2V) is rapidly hydrolyzed and converted to E2 during absorption in the gastrointestinal tract (1 mg of E2V contains 0.76 mg of E2).5 Piperazine estrone sulfate undergoes conversion to estrone after oral absorption.6 Therefore, a lab test for E1 and E2 will detect levels achieved following oral administration. Pregnant mares produce and excrete large volumes of conjugated estrogens in their urine: hence, the name Premarin for these conjugated equine estrogens (CEE). It is important to note that CEE represents a group of biologically active molecules rather than one specific compound and not a prodrug for estradiol. Although some conversion to E2 occurs, serum concentrations of E2 will not reflect the total estrogenicity. Ethinyl estradiol (EE), the form of estrogen used in most combined hormonal contraceptives (pills, patch, ring), also is not a prodrug for estradiol. Laboratory assays for E2 do not detect EE. Like estradiol, EE undergoes hepatic conjugation following oral administration, but unlike estradiol these conjugated forms remain highly potent, stimulating the liver on second pass. Therefore, even when delivered non-orally, EE exerts potent effects on the liver.
The effects of estrogens differ depending on dose, potency, molecule, and route of administration. Oral administration of estrogens exposes the gut and liver to high concentrations of steroid, resulting in extensive first-pass metabolism and hepatic stimulation.7 This results in the induction of important hepatic globulins involved in coagulation and in lipid and steroid hormone transport pathways.7 Of major interest is the induction of SHBG by estrogen. The increase in SHBG that occurs in users of combined hormonal contraceptive methods results in a decrease in free androgen levels, an important effect that provides the basis for treatment of androgen-related symptoms, such as acne and hirsutism. The degree of hepatic induction following oral administration of an oral estrogen is dose dependent.8,9
The most important clinical consequence of hepatic stimulation with estrogen therapy is the increased risk of venous (VTE) and arterial (ATE) thrombosis. Humans likely evolved this shift to clot formation in response to estrogen as an adaptation to reduce blood loss with pregnancy.10 Ordinarily, the ovary delivers estradiol directly into the circulation at physiologic levels. Placental production of E2 leads to an increase in the liver production of clotting proteins and an increased risk of blood clots during pregnancy. Numerous studies have documented the increase in thrombosis associated with estrogens used in contraception and menopausal therapy.
Estrogens are eliminated from the body by metabolic conversion to inactive molecules followed by excretion in the feces and urine. The first step in this metabolism requires hydroxylation catalyzed by cytochrome P450 (CYP) enzymes in the liver.11 Therefore, drugs that increase or decrease the activity of CYP enzymes will influence the level of circulating estrogens. CYP3A4 and CYP2C9 are the major isoforms contributing to the oxidative metabolism of EE in human liver microsomes.12 Thus, coadministration of drugs that induce or suppress CYP enzymes may affect the levels of estrogen used therapeutically.
Comparator studies evaluating the effect of ligands on the suppression of FSH and induction of hepatic globulins have demonstrated that the EE is about 100-fold more potent than E2.6 Since EE passes through the liver on first pass without extensive conjugation, the liver effects of EE remain potent on recirculation. To restate for emphasis, the enhanced effect of EE on induction of hepatic globulins occurs because of greater potency, lack of significant first- pass conjugation, and potent induction on recirculation. This leads to important clinical consequences. E2 undergoes isomerization to E1, a less potent estrogen. This is of particular importance following oral administration.13 Thus, the stimulation effects of E2 on the liver following oral administration occur primarily as a result of first-pass. Parenteral administration of E2 at physiologic levels does not result in significant hepatic stimulation.7 In contrast, EE induces potent stimulation regardless of route of administration.
In other words, while oral administration of any estrogen will exert dose- and potency-related increases in hepatic globulins, parenteral administration of natural estrogens at physiologic levels should not increase clot risk. In contrast, parenteral administration of EE does not reduce hepatic impact; hormonal contraception with transdermal or transvaginal administration of EE is associated with a risk of VTE similar to that observed with oral preparations.14
Evidence is accumulating that transdermal delivery of estradiol is not associated with an increased risk of VTE during postmenopausal hormonal therapy (HT). Canonico et al performed a case-control study that evaluated the associations of obesity and estrogen use on VTE risk in postmenopausal women in France. They found an increased risk for overweight (adjusted odds ratio [aOR], 2.5; 95% confidence interval [CI], 1.7-3.7; and aOR, 3.9; 95% CI, 2.2-6.9) for obese women compared to normal weight. They also reported that oral estrogen increased the risk four-fold (aOR, 4.5; 95% CI, 2.6-7.7), but found no increase in women using transdermal E2 (aOR, 1.1; 95% CI, 0.7-1.7) compared to nonusers. They reported significant interaction with the combination of oral estrogen use and overweight or obesity (aOR, 10.2; 95% CI, 3.5-30.2; aOR, 20.6; 95% CI, 4.8-88.1, respectively), but no excess risk among overweight and obese users of transdermal E2.15
A 2019 publication from the United Kingdom has provided additional evidence supporting these results. Using data from the UK Clinical Practice Research Datalink databases, Vinogradova et al completed a case-control study that compared 80,396 women between 40 and 79 years of age with a primary diagnosis of venous thromboembolism between 1998 and 2017 to a group of 391,494 female controls matched by age, general practice, and index date. They calculated odds ratios and adjusted these for demographics, smoking status, alcohol consumption, comorbidities, recent medical events, and other prescribed drugs. They found an overall increase in VTE with any hormone therapy consistent with other studies, but the risk was confined to oral therapy. Transdermal E2 therapy did not increase the risk of VTE (aOR, 0.93; 95% CI, 0.87-1.01).16 These results are highly encouraging and are consistent with our biologic understanding of the effects of estrogens on the liver.
These results should inform our use of estrogen therapy in postmenopausal women. Transdermal or vaginal delivery of estradiol should increase cardiovascular safety. Although we do not have evidence that parenteral estradiol reduces the risk of heart attack and stroke, I predict these data will follow.
What about hormonal contraception? Over the last decade, two oral contraceptives based on estradiol have been introduced: E2/nomegestrol acetate (not available in the United States) and E2V/dienogest vaginal rings containing estradiol.17 A combination of E4 with drospirenone pill18 is in clinical trials. The estradiol ring is particularly exciting, as we should expect results similar to those observed with transdermal estrogen therapy in postmenopausal women. Estradiol pills still provide hepatic stimulation on first pass, but the effect on hepatic globulin synthesis appears lower than EE.19,20 While interesting, we cannot rely on unvalidated surrogate markers to predict a clinical benefit.
The INAS-SCORE study investigated the cardiovascular risks associated with the use of the E2V/dienogest pill in women from the United States and Europe.21 Investigators enrolled 50,203 new COC users, and completed up to 5.5 years of follow-up. Overall, 20% of the cohort used the E2 pill and 80% used EE pills. Compared to all EE COCs, the adjusted hazard ratios for DVT for the E2V/DNG was 0.5 (95% CI, 0.2-1.0). Although the point estimate suggesting a lower risk is intriguing, the confidence interval overlaps 1.0 supporting caution in our enthusiasm.
Estetrol, also appears to have lower impact on hemostatic factors following oral administration.22 Confirming whether oral contraceptives currently in development using Estetrol or estradiol-containing vaginal rings will have less thrombosis risk than existing CHC products will require similar population-based trials. This new emphasis on approaches to estrogen administration that reduce hepatic stimulation should improve safety of combined hormonal contraception. In my opinion, this represents a major breakthrough. While the differential impact of various progestins have preoccupied us for many years, the overall effect on thrombosis risk appears minimal. In contrast, moving away from ethinyl estradiol should improve outcomes and reduce risk.
- Hammond GL. Plasma steroid-binding proteins: Primary gatekeepers of steroid hormone action. J Endocrinol 2016;230:R13-R25.
- Fritz MA, Speroff L. Clinical Gynecologic Endocrinology and Infertility, 8th Ed. Philadelphia: Lippincott Williams & Wilkins; 2011.
- Blackburn ST. Maternal, Fetal, & Neonatal Physiology: A Clinical Perspective. Amsterdam: Elsevier; 2017.
- Coelingh Bennink HJ, et al. Estetrol review: Profile and potential clinical applications. Climacteric 2008;11(Suppl 1):47-58.
- Jensen JT. Evaluation of a new estradiol oral contraceptive: Estradiol valerate and dienogest. Expert Opin Pharmacother 2010;11:1147-1157.
- Mashchak CA, et al. Comparison of pharmacodynamic properties of various estrogen formulations. Am J Obstet Gynecol 1982;144:511-518.
- Balfour JA, Heel RC. Transdermal estradiol. A review of its pharmacodynamic and pharmacokinetic properties, and therapeutic efficacy in the treatment of menopausal complaints. Drugs 1990;40:561-582.
- Ottosson UB, et al. Estrogen induction of liver proteins and high-density lipoprotein cholesterol: Comparison between estradiol valerate and ethinyl estradiol. Gynecol Obstet Invest 1986;22:198-205.
- Matsui S, et al. Sex hormone-binding globulin and antithrombin III activity in women with oral ultra-low-dose estradiol. J Obstet Gynaecol 2017;37:627-632.
- ESHRE Capri Workshop Group. Venous thromboembolism in women: A specific reproductive health risk. Hum Reprod Update 2013;19:471-482.
- Tsuchiya Y, et al. Cytochrome P450-mediated metabolism of estrogens and its regulation in human. Cancer Lett 2005;227:115-124.
- Wang B, et al. The involvement of CYP3A4 and CYP2C9 in the metabolism of 17 alpha-ethinylestradiol. Drug Metab Dispos 2004;32:1209-1212.
- Kopper NW, et al. Transdermal hormone therapy in postmenopausal women: A review of metabolic effects and drug delivery technologies. Drug Des Devel Ther 2009;2:193-202.
- Dinger J, et al. Cardiovascular risk associated with the use of an etonogestrel-containing vaginal ring. Obstet Gynecol 2013;122:800-808.
- Canonico M, et al. Obesity and risk of venous thromboembolism among postmenopausal women: Differential impact of hormone therapy by route of estrogen administration. The ESTHER Study. J Thromb Haemost 2006;4:1259-1265.
- Vinogradova Y, et al. Use of hormone replacement therapy and risk of venous thromboembolism: Nested case-control studies using the QResearch and CPRD databases. BMJ 2019;364:k4810.
- Jensen JT, et al. Continuous dosing of a novel contraceptive vaginal ring releasing Nestorone and estradiol: Pharmacokinetics from a dose-finding study. Contraception 2018;97:422-427.
- Duijkers IJ, et al. Inhibition of ovulation by administration of estetrol in combination with drospirenone or levonorgestrel: Results of a phase II dose-finding pilot study. Eur J Contracept Reprod Health Care 2015;20:476-489.
- Raps M, et al. Resistance to APC and SHBG levels during use of a four-phasic oral contraceptive containing dienogest and estradiol valerate: A randomized controlled trial. J Thromb Haemost 2013;11:855-861.
- Klipping C, et al. Hemostatic effects of a novel estradiol-based oral contraceptive: An open-label, randomized, crossover study of estradiol valerate/dienogest versus ethinylestradiol/levonorgestrel. Drugs R D 2011;11:159-170.
- Dinger J, et al. Impact of estrogen type on cardiovascular safety of combined oral contraceptives. Contraception 2016;94:328-339.
- Farris M, et al. Pharmacodynamics of combined estrogen-progestin oral contraceptives: 2. Effects on hemostasis. Expert Rev Clin Pharmacol 2017;10:1129-1144.