Estrogens and progestins are among the most widely prescribed drugs in clinical medicine, applied across an extraordinary range of indications that include hormonal contraception, menopausal hormone therapy, endometriosis, uterine fibroids, ovulation induction, and gender-affirming care. Rational prescribing in each of these contexts requires a detailed understanding of receptor pharmacology, estrogen and progestin biosynthesis, the comparative pharmacokinetics of available agents and formulations, and a working knowledge of the clinically significant drug interactions that can render hormonal therapy ineffective or dangerous. This module establishes the pharmacological foundation for the entire gonadal pharmacology series by addressing receptor subtypes and their tissue-selective signaling consequences, the biosynthetic pathways that produce endogenous estrogens and progestins, the pharmacological profiles of natural, synthetic, and conjugated estrogen preparations and the spectrum of synthetic progestins, and the absorption, distribution, metabolism, and excretion characteristics that govern clinical formulation selection.
The biological effects of estrogens are mediated primarily through two nuclear receptor subtypes, estrogen receptor alpha (ERα, encoded by ESR1) and estrogen receptor beta (ERβ, encoded by ESR2), which belong to the nuclear receptor superfamily of ligand-activated transcription factors. Both receptors share a conserved domain architecture consisting of an N-terminal activation function 1 (AF-1) domain, a central deoxyribonucleic acid (DNA)-binding domain (DBD) with two zinc finger motifs, a hinge region, and a C-terminal ligand-binding domain (LBD) that contains the activation function 2 (AF-2) surface required for coactivator recruitment. Despite sharing approximately 97% amino acid identity in the DBD, the LBDs of ERα and ERβ share only about 59% identity, which creates differential ligand-binding affinity and explains why selective agonists and antagonists with tissue-selective profiles are pharmacologically achievable.1
ERα is the dominant receptor subtype in the uterus, liver, breast, bone (osteoblasts), cardiovascular endothelium, and the hypothalamic-pituitary axis where it mediates the negative feedback of estradiol on gonadotropin-releasing hormone (GnRH) and luteinizing hormone (LH) secretion. ERβ is expressed preferentially in the ovary (granulosa cells), colon, lung, prostate, and central nervous system, and its activation generally opposes or modulates ERα-driven proliferative effects in tissues where both are co-expressed, including the breast and endometrium. This opposing functional relationship between the two subtypes is the conceptual basis for selective estrogen receptor modulators (SERMs), which exploit differences in receptor conformation induced by different ligands to produce tissue-selective agonism and antagonism.12
The classical genomic signaling pathway proceeds through ligand binding to the ER LBD, which induces a conformational change that dissociates heat shock proteins (HSP90, HSP70), allows receptor dimerization (ERα–ERα, ERβ–ERβ, or ERα–ERβ heterodimers), and facilitates binding of the dimer to estrogen response elements (EREs) in the promoter regions of target genes. Coactivator complexes, including steroid receptor coactivators (SRC-1, SRC-2, SRC-3) of the p160 family and the mediator complex, are recruited through the AF-2 surface and activate transcription of estrogen-responsive genes. The genomic pathway operates on a timescale of hours because it requires messenger ribonucleic acid (RNA) synthesis and protein translation. Non-genomic estrogen signaling occurs through a membrane-associated pool of ER and through the G protein-coupled estrogen receptor (GPER), formerly known as GPR30 [G protein-coupled receptor 30], and produces rapid effects within seconds to minutes, including activation of adenylyl cyclase, phospholipase C, and the mitogen-activated protein kinase (MAPK) and phosphoinositide 3-kinase (PI3K)–Akt pathways. These rapid non-genomic effects are clinically relevant because they mediate estrogen-dependent vasodilation through endothelial nitric oxide synthase (eNOS) activation, which partially explains the cardiovascular protection observed with estrogen in young premenopausal women.3
Progesterone receptors also exist in two principal isoforms, progesterone receptor A (PR-A) and progesterone receptor B (PR-B), both encoded by the same gene (PGR) through use of different promoters and translational start sites. PR-B is the full-length receptor and a transcriptional activator in most reproductive tissues; PR-A lacks the first 164 amino acids of PR-B and functions primarily as a transcriptional repressor, capable of inhibiting PR-B as well as other nuclear receptors including ERα. The PR-A/PR-B (progesterone receptor A/B) ratio varies across tissues and across reproductive cycle phases and is altered by estrogen priming, which upregulates both isoforms. Synthetic progestins differ from natural progesterone in their relative affinity for PR-A versus PR-B, and they also exhibit variable cross-reactivity with the androgen receptor (AR), glucocorticoid receptor (GR), and mineralocorticoid receptor (MR), which is the molecular basis for the androgenic, anti-androgenic, and mineralocorticoid side effect profiles that distinguish progestin generations from one another.4
The tissue-selective consequences of ER subtype distribution underpin SERM pharmacology. Tamoxifen acts as an ERα partial agonist in bone and endometrium but an antagonist in breast tissue. Raloxifene acts as an ERα antagonist in both breast and endometrium while retaining agonism in bone. These differences are determined by the specific conformational change each ligand induces in the ER LBD, which in turn determines which coactivators are recruited. Understanding receptor subtype distribution by tissue is essential for predicting tissue-specific drug effects throughout this chapter series.
Ovarian estrogen biosynthesis is organized around the two-cell, two-gonadotropin model in which luteinizing hormone (LH) and follicle-stimulating hormone (FSH) act on distinct cellular compartments to produce estradiol (E2) through cooperative steroidogenesis. LH acts on theca interna cells via the LH receptor (LHR), a Gs-coupled receptor that activates adenylyl cyclase and raises intracellular cyclic AMP (cAMP), stimulating protein kinase A (PKA)-dependent phosphorylation of the steroidogenic acute regulatory protein (StAR). StAR mediates the rate-limiting translocation of cholesterol from the outer to the inner mitochondrial membrane, where the cytochrome P450 side-chain cleavage enzyme (CYP11A1) converts cholesterol to pregnenolone. Pregnenolone is then converted to progesterone by 3β-hydroxysteroid dehydrogenase/Δ5-Δ4 isomerase (3β-HSD), and progesterone is sequentially hydroxylated by CYP17A1 (cytochrome P450 17A1, the 17α-hydroxylase/lyase enzyme) to 17α-hydroxyprogesterone, then converted to androstenedione by the lyase activity of CYP17A1. Androstenedione and testosterone, which are androgens, are the substrates exported from the theca cell into the granulosa cell.5
FSH acts on granulosa cells via the FSH receptor (FSHR), which also signals through Gs-cAMP-PKA and simultaneously upregulates aromatase (CYP19A1), the enzyme responsible for converting theca-derived androgens (androstenedione and testosterone) to estrogens (estrone and estradiol, respectively). Aromatase catalyzes three sequential hydroxylation reactions on the A ring of the androgen substrate, with loss of the C-19 methyl group and aromatization of the A ring, yielding the phenolic A ring characteristic of all estrogens. Estradiol (E2) is the primary secretory product of the preovulatory follicle, with plasma levels rising steeply in the late follicular phase to concentrations above 200 picograms per milliliter, triggering the positive feedback LH surge that induces ovulation. Following ovulation, the granulosa-lutein cells of the corpus luteum produce both estradiol and large amounts of progesterone under continued LH stimulation.56
Three naturally occurring estrogens are present in women: estradiol (E2), estrone (E1), and estriol (E3). Estradiol is the most potent, produced primarily in the ovary during reproductive years, with a relative binding affinity for ERα approximately ten-fold greater than estrone. Estrone is produced mainly by peripheral aromatization of adrenal androstenedione in adipose tissue, liver, and muscle, and becomes the dominant circulating estrogen after menopause since ovarian production ceases. Estriol is a weak estrogen produced in large quantities by the placenta during pregnancy through fetal adrenal and hepatic metabolism of dehydroepiandrosterone sulfate (DHEA-S); its high levels during pregnancy are responsible for several pregnancy-specific estrogenic effects including vaginal epithelial proliferation, and its low systemic potency relative to estradiol is clinically exploited in local vaginal preparations. The hypothalamic-pituitary-ovarian (HPO) axis exerts negative feedback control over FSH and LH secretion through estradiol acting on ERα in the hypothalamus and pituitary; the mid-cycle switch from negative to positive feedback, mediated by sustained high estradiol concentrations, is a unique feature of the female HPO axis and triggers the preovulatory LH surge.5
After menopause, ovarian follicular function ceases and circulating estradiol falls to levels below 20 picograms per milliliter. The dominant postmenopausal estrogen becomes estrone, derived from peripheral aromatization of adrenal androstenedione in adipose tissue. Because aromatase expression in adipose tissue is not gonadotropin-dependent, postmenopausal estrone production correlates positively with body mass index (BMI), which explains in part why obese postmenopausal women retain more estrogen exposure and have both lower rates of menopausal vasomotor symptoms and higher rates of estrogen receptor-positive breast cancer and endometrial cancer compared to lean postmenopausal women. The therapeutic implication is that aromatase inhibitors, which block CYP19A1, are highly effective in postmenopausal women with estrogen receptor-positive breast cancer because they suppress the only significant remaining source of estrogen in these patients.6
The theca cell makes androgens (LH-driven); the granulosa cell converts them to estrogens (FSH-driven via aromatase). This model predicts several clinical phenomena: polycystic ovary syndrome (PCOS) involves excess theca androgen production with impaired granulosa aromatization; aromatase inhibitors and FSH-based ovulation induction both target discrete steps in this pathway; and aromatase inhibitors suppress estrogen in postmenopausal women because peripheral adipose aromatase, not the ovary, is the remaining estrogen source. This model reappears throughout Ova-03 and Ova-04.
Exogenous estrogen preparations span a wide pharmacological spectrum from bioidentical 17β-estradiol to highly potent synthetic derivatives designed to resist hepatic first-pass degradation. The clinical choice of estrogen agent is not pharmacologically neutral: different preparations produce substantially different hepatic effects, binding affinities, and metabolic consequences that are directly relevant to venous thromboembolism (VTE) risk, lipid profiles, and drug interactions. Understanding the pharmacological distinctions between estrogen types is prerequisite to making evidence-based formulation decisions in hormone therapy and contraception.7
17β-Estradiol is the bioidentical form of the primary ovarian estrogen and is available in oral, transdermal, vaginal, injectable, and implantable formulations. When administered orally, estradiol undergoes extensive first-pass metabolism in the intestinal mucosa and liver, where it is converted to estrone by 17β-hydroxysteroid dehydrogenase and then to estrone sulfate and estrone glucuronide, resulting in an oral bioavailability of approximately 5% and a plasma estrone-to-estradiol ratio that greatly exceeds the physiological ratio seen with endogenous ovarian secretion. This first-pass conversion is clinically significant because the liver is exposed to supraphysiological estrogen concentrations through the portal circulation, stimulating hepatic synthesis of sex hormone-binding globulin (SHBG), angiotensinogen, coagulation factors (factors VII, IX, X, fibrinogen), and C-reactive protein (CRP) at levels that are not observed with transdermal estradiol, which bypasses portal hepatic first-pass. The oral estradiol dose required to achieve systemic efficacy (typically 1–2 mg/day) therefore carries a hepatic amplification effect not shared by transdermal preparations, which deliver estradiol directly into the systemic circulation at doses of 25–100 micrograms per 24 hours.78
Ethinyl estradiol (EE) is a synthetic 17α-ethynyl derivative of estradiol that was developed specifically for oral use. The 17α-ethynyl group prevents oxidative metabolism at the C-17 position by cytochrome P450 3A4 (CYP3A4), conferring resistance to first-pass hepatic degradation and resulting in oral bioavailability of approximately 40–45%, far superior to oral estradiol. Despite this improvement in bioavailability, EE is still subject to intestinal metabolism by CYP3A4 and sulfotransferases, and its bioavailability is highly variable across individuals (range 20–65%), partly due to polymorphisms in intestinal CYP3A4 expression. EE is considerably more potent than estradiol in stimulating hepatic protein synthesis because its 17α-ethynyl group also impairs hepatic inactivation via 17β-HSD, resulting in prolonged hepatic receptor activation. Consequently, even the low doses of EE used in modern combined oral contraceptives (20–35 micrograms per day) produce substantially greater hepatic estrogenic stimulation than physiologically equivalent doses of transdermal estradiol. This hepatic potency of EE is the mechanistic basis for the elevated VTE risk associated with combined oral contraceptive use and is clinically distinguishable from the VTE risk profile of transdermal hormone therapy.79
CEE (conjugated equine estrogens, brand name Premarin) consist of a complex mixture of water-soluble estrogen sulfates extracted from the urine of pregnant mares. The major components are sodium estrone sulfate (approximately 50–60%), sodium equilin sulfate (approximately 22–30%), and smaller amounts of 17α-dihydroequilin, 17β-dihydroequilin, equilenin, and delta-8,9-dehydroestrone sulfate. The equine estrogens (equilin, equilenin) have binding affinities for ERα comparable to estradiol but have considerably longer half-lives because they are poorly converted to inactive metabolites by standard human hepatic enzymes; equilin sulfate, in particular, accumulates in adipose tissue during prolonged use and may be detectable in plasma for weeks after discontinuation of CEE. This prolonged biological activity is not shared by estradiol formulations and has implications for interpreting the Women's Health Initiative (WHI) CEE data in relation to preparations used in European clinical practice, which predominantly used transdermal estradiol rather than CEE.8
Estetrol (E4) is a native fetal estrogen produced by the fetal liver from estradiol via sequential 15α- and 16α-hydroxylation and available in a combined oral contraceptive formulation paired with drospirenone (Nexstellis/Drovelis). Estetrol acts as a selective estrogen receptor modulator with tissue-specific agonist/antagonist activity defined by its ER-selective agonism: it binds ERα and ERβ with lower affinity than estradiol but exhibits limited coactivator recruitment in breast tissue while retaining contraceptive efficacy through hypothalamic-pituitary suppression. It does not bind to GPER (G protein-coupled estrogen receptor) and has minimal activation of hepatic estrogenic signaling, resulting in a more favorable hepatic impact (lower SHBG increase, lower CRP elevation) compared to EE-containing formulations, though long-term VTE and cardiovascular outcome data remain limited compared to the established EE evidence base.7
The hepatic amplification effect of oral EE vs. transdermal estradiol is clinically decisive for VTE risk. Oral EE in combined oral contraceptives increases VTE risk approximately 3–4-fold above baseline. Transdermal estradiol in hormone therapy preparations does not appear to increase VTE risk in observational data. When prescribing estrogen to patients with risk factors for VTE (obesity, personal or family history, thrombophilia), the route of administration matters as much as the dose. This is covered in detail in Ova-03.
Natural progesterone binds with high affinity and selectivity to the progesterone receptor (PR) and has low affinity for the androgen receptor (AR), glucocorticoid receptor (GR), and mineralocorticoid receptor (MR). Its clinical utility as an oral agent is severely limited by rapid first-pass metabolism: oral bioavailability of crystalline progesterone is less than 10%, and peak plasma levels are highly variable. Micronized progesterone (particle size reduced to 10–50 microns, suspended in oil-filled capsules, brand name Prometrium and Utrogestan) achieves substantially better oral bioavailability of approximately 10–15%, sufficient for endometrial protection in hormone therapy and luteal phase support in assisted reproduction, and is metabolized by cytochrome P450 3A4 (CYP3A4) and 5α-reductase to allopregnanolone and pregnanolone, active 5α-reduced metabolites that are positive allosteric modulators of the gamma-aminobutyric acid type A (GABAA) receptor and produce the sedative and anxiolytic side effects characteristic of oral micronized progesterone. Intramuscular and vaginal progesterone formulations bypass first-pass metabolism and achieve reliable therapeutic levels for specific indications such as luteal phase support in in vitro fertilization (IVF) cycles.10
Synthetic progestins were developed to overcome the bioavailability limitations of natural progesterone and are classified into generations based on their structural derivation and their pharmacological receptor selectivity profiles. First-generation progestins derived from 17α-hydroxyprogesterone include medroxyprogesterone acetate (MPA) and megestrol acetate. MPA has high progestational activity, modest glucocorticoid receptor affinity, and negligible androgenic activity, but it does not share the GABAA-potentiating metabolites of natural progesterone and differs meaningfully from micronized progesterone in clinical outcome data, including a differential breast cancer signal in the Women's Health Initiative (WHI) trial. Progestins derived from 19-nortestosterone include the norethindrone generation (first-generation 19-nor progestins: norethindrone, norethindrone acetate, norethynodrel, ethynodiol diacetate) and the levonorgestrel generation (second-generation 19-nor progestins: levonorgestrel, norgestrel). All 19-nortestosterone-derived progestins retain structural similarity to testosterone and have measurable AR binding affinity that is substantially higher than natural progesterone; levonorgestrel in particular has the highest androgenic index among the commonly used progestins, which is clinically expressed as adverse effects on the lipid profile (reduction in high-density lipoprotein cholesterol) and androgenic skin effects when used at higher doses.4
Third-generation progestins (desogestrel, gestodene, norgestimate) were developed to reduce the androgenic activity of second-generation 19-nor progestins. These agents have higher PR binding affinity with substantially reduced AR activity, resulting in a less adverse lipid profile compared to levonorgestrel-containing formulations. However, third-generation progestins are associated with a higher venous thromboembolism (VTE) risk than second-generation progestins in combined oral contraceptive formulations, an observation that has generated ongoing debate but is supported by multiple epidemiological studies and attributed in part to differential sex hormone-binding globulin (SHBG) induction and to specific progestin-related effects on coagulation independent of their androgenicity. Desogestrel is the precursor to etonogestrel (3-ketodesogestrel), the active metabolite that is the progestin component of the subdermal implant (Nexplanon) and the vaginal ring (NuvaRing/Annovera).11
Fourth-generation progestins include drospirenone, dienogest, nomegestrol acetate, and trimegestone. Drospirenone is derived from spironolactone and is unique among progestins for having anti-androgenic activity through AR antagonism and anti-mineralocorticoid activity through MR antagonism, the latter producing mild natriuretic and blood pressure-lowering effects. The anti-mineralocorticoid activity of drospirenone is clinically significant: it counteracts the sodium retention associated with the estrogenic component of combined oral contraceptives and is the basis for drospirenone-containing pills being used for premenstrual dysphoric disorder (PMDD) and mild hypertension-related contraceptive selection. However, its MR antagonism also raises the risk of hyperkalemia in patients taking other potassium-retaining agents (angiotensin-converting enzyme [ACE] inhibitors, angiotensin receptor blockers (ARBs), potassium-sparing diuretics, NSAIDs), and serum potassium monitoring is recommended in these patients. Dienogest is a 19-nor progestin with strong PR agonism, anti-androgenic activity comparable to cyproterone acetate, and minimal GR, MR, and AR cross-reactivity; it is used in combined oral contraceptives paired with estradiol valerate (Natazia/Qlaira) and is approved for endometriosis in progestin-only preparations.4
Androgenic progestins (levonorgestrel, norgestrel, norethindrone acetate) worsen lipid profiles and may cause acne and hirsutism at higher doses. Anti-androgenic progestins (drospirenone, dienogest, cyproterone acetate) are appropriate for patients with androgen-excess conditions (PCOS, hirsutism) but require potassium monitoring with drospirenone if concomitant potassium-retaining drugs are used. MPA differs from micronized progesterone in breast cancer and neurological outcomes data and should not be considered pharmacologically equivalent. These distinctions recur throughout Ova-02 and Ova-03.
The pharmacokinetics of estrogens and progestins are profoundly route-dependent, and selecting the appropriate formulation requires understanding how each route determines the plasma concentration-time profile, the degree of hepatic first-pass exposure, and the resulting systemic and hepatic pharmacological consequences. Oral estradiol undergoes extensive first-pass extraction in the small intestinal mucosa and liver, resulting in a predominantly estrone-rich systemic circulation rather than an estradiol-rich profile; the oral estradiol dose must be three to five times the transdermal dose to achieve equivalent systemic estradiol concentrations. The terminal half-life of oral estradiol after hepatic metabolism is 12–24 hours, but the biologically active estrone sulfate pool that accumulates acts as a reservoir extending estrogenic activity. Oral ethinyl estradiol (EE) is metabolized by intestinal and hepatic cytochrome P450 3A4 (CYP3A4) and cytochrome P450 2C9 (CYP2C9) to 2-hydroxy-EE and other catechol estrogens, with a terminal half-life of approximately 24–36 hours and a pronounced hepatic first-pass effect that, despite its magnitude, still leaves sufficient intact EE to produce strong hepatic estrogenic stimulation.7
Transdermal estradiol patches, gels, and sprays deliver estradiol directly through the skin into the systemic circulation, bypassing portal hepatic first-pass. The absorbed estradiol enters the systemic venous circulation and reaches the liver at concentrations proportional to systemic plasma levels rather than at the concentrated portal bolus concentrations produced by oral administration. As a result, transdermal estradiol at therapeutically effective doses (25–100 micrograms per 24 hours for patches; 0.5–1 mg per day for gels) produces minimal increases in hepatic sex hormone-binding globulin (SHBG), C-reactive protein (CRP), angiotensinogen, and coagulation factors compared to oral estradiol, and observational studies including the ESTHER (Estrogen and Thromboembolism Risk) study and meta-analyses by Canonico et al. consistently demonstrate no increase in venous thromboembolism (VTE) risk with transdermal estradiol even at higher doses, a finding that strongly supports preferential use of the transdermal route in women with VTE risk factors.8
Vaginal estrogen preparations (cream, suppository, ring) produce primarily local mucosal and submucosal effects with minimal systemic absorption at standard low doses (0.5–2 grams of 0.01% estradiol cream delivering 50–100 micrograms per application; the low-dose 10-microgram estradiol vaginal insert delivers serum estradiol levels near the postmenopausal baseline). These preparations are used for genitourinary syndrome of menopause (GSM), formerly called vulvovaginal atrophy, and are generally safe in breast cancer survivors when systemic absorption is minimal, though oncological guidance varies by individual risk. Injectable estrogen formulations, including estradiol cypionate (oil suspension, intramuscular) and estradiol valerate (oil, intramuscular), provide sustained release over 1–4 weeks with wide peak-to-trough variation; estradiol valerate is also available in an oral combined pill paired with dienogest (Qlaira/Natazia) where the ester is hydrolyzed rapidly after absorption to release estradiol, minimizing the hepatic first-pass potency difference relative to EE, though a degree of first-pass exposure still occurs.7
Progestin pharmacokinetics are similarly route-dependent. Oral medroxyprogesterone acetate (MPA) is well absorbed with approximately 90% bioavailability and a half-life of 24–30 hours, reaching steady-state within 3–5 days. It is metabolized by CYP3A4 and stored in adipose tissue, producing a depot effect with prolonged measurable plasma levels after cessation, a property exploited in the DMPA (depot medroxyprogesterone acetate) injectable formulation (150 mg intramuscular every 12 weeks), which achieves contraceptive serum levels for 3 months while suppressing ovulation within 24 hours of injection. The DMPA pharmacokinetic profile results in a delayed return to fertility averaging 9–10 months after the last injection, a clinically relevant counseling point. Levonorgestrel is orally bioavailable (approximately 90%), highly protein-bound to SHBG (approximately 50%) and albumin (approximately 47%), with a free fraction of only 1–3%; because SHBG is induced by EE in combined pills, coadministration with EE reduces the free levonorgestrel fraction, which is one reason levonorgestrel dose requirements differ between combined and progestin-only preparations.411
Etonogestrel (the active metabolite of desogestrel) achieves contraceptive plasma concentrations within 8 hours of subdermal implant insertion (Nexplanon), with initial levels of approximately 400–600 picograms per milliliter falling to approximately 180–200 picograms per milliliter by year 3, which remains above the threshold for ovulation suppression (approximately 90 picograms per milliliter) throughout the 3-year approved duration. Etonogestrel is metabolized by CYP3A4 to inactive hydroxylated metabolites and has a terminal half-life of approximately 25 hours after implant removal, allowing return of ovulation within 3–4 weeks. The vaginal contraceptive ring (NuvaRing) releases both EE (15 micrograms per day) and etonogestrel (120 micrograms per day) directly through the vaginal mucosa, achieving systemic concentrations comparable to low-dose combined oral contraceptives but with reduced peak hepatic EE exposure relative to oral administration because vaginal absorption bypasses the intestinal first-pass component, though portal hepatic passage still occurs once absorbed into the systemic circulation.9
Both estradiol and testosterone circulate largely bound to SHBG (high affinity) and albumin (lower affinity, higher capacity); only the free fraction is biologically active. EE in combined oral contraceptives markedly induces hepatic SHBG production, raising SHBG levels 2–4-fold and reducing free testosterone, which is the basis for combined oral contraceptive use in androgen-excess conditions (PCOS, hirsutism). Progestins with androgenic activity (levonorgestrel) compete with testosterone for SHBG binding, increasing free testosterone; anti-androgenic progestins (drospirenone) do not. SHBG levels therefore serve as a pharmacodynamic marker of hepatic estrogenic stimulation and as a determinant of free androgen bioavailability.
Drug interactions involving estrogen and progestin preparations are predominantly cytochrome P450 3A4 (CYP3A4)-mediated, reflecting the central role of this enzyme in estrogen and progestin oxidative metabolism. Because both ethinyl estradiol (EE) and most synthetic progestins are CYP3A4 substrates, any agent that induces CYP3A4 or intestinal P-glycoprotein (P-gp) will accelerate their metabolism, lower plasma concentrations, and potentially result in contraceptive failure or reduced therapeutic efficacy. The clinical consequence of this interaction ranges from minor reductions in hormone levels to complete loss of ovulation suppression, depending on the potency of the inducer. Conversely, CYP3A4 inhibitors raise EE and progestin levels, potentially increasing estrogen-related adverse effects including nausea, breast tenderness, and cycle spotting.13
Rifampin (rifampicin) is the most potent CYP3A4 inducer encountered in clinical practice and produces the most clinically significant interaction with hormonal contraceptives. Rifampin induces both CYP3A4 and cytochrome P450 2C9 (CYP2C9), reduces EE plasma area under the curve (AUC) by greater than 50%, and reduces progestin AUCs comparably. Even short courses of rifampin (7–14 days) produce induction that persists for 4–6 weeks after cessation due to the time required for CYP3A4 enzyme turnover to return to baseline. Patients on combined oral contraceptives requiring rifampin treatment for tuberculosis or rifabutin for Mycobacterium avium complex prophylaxis must be advised that hormonal contraception is unreliable during and for at least 4 weeks after rifampin therapy and should use barrier contraception or a copper intrauterine device (IUD) as a non-hormonal alternative. Rifabutin is a weaker CYP3A4 inducer and a clinically relevant interaction is less certain but still supported by available data.12
Among antiepileptic drugs (AEDs), enzyme-inducing agents that reduce hormonal contraceptive efficacy include carbamazepine, phenytoin, phenobarbital, primidone, oxcarbazepine, rufinamide, and topiramate (at doses above 200 mg per day). The World Health Organization Medical Eligibility Criteria for Contraceptive Use (WHO MEC) classifies combined hormonal contraceptives and progestin-only pills as Category 3 (risks generally outweigh advantages) and the levonorgestrel intrauterine device (IUD), depot medroxyprogesterone acetate (DMPA), and implant as Category 2 or 3 depending on the specific inducing antiepileptic drug (AED), because the local endometrial effect of the levonorgestrel-intrauterine device (LNG-IUD) and the depot serum levels of DMPA are considered more robust to induction than oral preparations. Non-enzyme-inducing AEDs that do not affect hormonal contraceptive efficacy include valproate, levetiracetam, lamotrigine, gabapentin, pregabalin, and vigabatrin; patients on these agents can use any hormonal method without interaction concerns related to enzyme induction.13
The lamotrigine–EE interaction is clinically distinctive and bidirectional. EE potently induces the uridine diphosphate (UDP)-glucuronosyltransferase enzyme UGT1A4 [uridine diphosphate-glucuronosyltransferase 1A4], which is the primary metabolic pathway for lamotrigine glucuronidation and inactivation. In women taking lamotrigine for epilepsy who start a combined oral contraceptive containing EE, lamotrigine plasma concentrations fall by approximately 40–65% within the first weeks of combined pill use, substantially increasing the risk of breakthrough seizures. Conversely, when the combined pill is stopped or during the pill-free interval of cyclic regimens, lamotrigine levels rebound sharply and may precipitate lamotrigine toxicity (dizziness, diplopia, ataxia). This interaction does not occur with progestin-only methods, which do not contain EE and do not induce UGT1A4. Management requires dose adjustment of lamotrigine at initiation and cessation of the combined pill, ideally in collaboration with the prescribing neurologist, and strongly favors progestin-only or non-hormonal contraception in women with epilepsy who are stabilized on lamotrigine.13
Human immunodeficiency virus (HIV) antiretrovirals represent another major class of CYP3A4 inducers affecting hormonal contraceptive efficacy. Non-nucleoside reverse transcriptase inhibitors (NNRTIs) vary: efavirenz and nevirapine are potent CYP3A4 inducers that substantially reduce EE and progestin levels; etravirine is a moderate inducer; rilpivirine does not induce CYP3A4 and does not impair contraceptive efficacy. Ritonavir-boosted protease inhibitors present a complex interaction: ritonavir is a potent CYP3A4 inhibitor at most substrates but paradoxically induces EE glucuronidation, resulting in net reduction of EE AUC by approximately 40–50% with ritonavir-boosted lopinavir and similar regimens. Integrase strand-transfer inhibitors (INSTIs) including dolutegravir, raltegravir, and bictegravir do not have clinically significant interactions with hormonal contraceptives and are preferred antiretroviral agents in women requiring reliable hormonal contraception. CYP3A4 inhibitors including azole antifungals (fluconazole, itraconazole, ketoconazole) raise EE levels and may increase estrogen-related adverse effects, though this interaction is generally clinically manageable and does not represent a contraindication to combined hormonal methods.13
No hormonal oral, patch, or ring contraceptive method should be considered reliable during rifampin therapy or for 4 weeks after its cessation. The copper IUD is the most appropriate emergency backup for patients needing rifampin who require ongoing contraception. The lamotrigine interaction is the most common clinically encountered bidirectional interaction: combined pill initiation reduces lamotrigine by up to 65% (seizure risk) and cessation raises it sharply (toxicity risk). Progestin-only methods avoid both problems and are the preferred hormonal option in lamotrigine-treated epilepsy.
Heldring N, Pike A, Andersson S, et al. Estrogen receptors: how do they signal and what are their targets. Physiol Rev. 2007;87(3):905–931.
doi:10.1152/physrev.00026.2006Nilsson S, Makela S, Treuter E, et al. Mechanisms of estrogen action. Physiol Rev. 2001;81(4):1535–1565.
doi:10.1152/physrev.2001.81.4.1535Levin ER. Membrane oestrogen receptor alpha signalling to cell functions. J Physiol. 2009;587(21):5019–5023.
doi:10.1113/jphysiol.2009.177097Schindler AE, Campagnoli C, Druckmann R, et al. Classification and pharmacology of progestins. Maturitas. 2008;61(1–2):171–180.
doi:10.1016/j.maturitas.2008.11.013Drummond AE, Fuller PJ. The importance of ERbeta signalling in the ovary. J Endocrinol. 2010;205(1):15–23.
doi:10.1677/JOE-09-0379Simpson ER, Davis SR. Minireview: aromatase and the regulation of estrogen biosynthesis — some new perspectives. Endocrinology. 2001;142(11):4589–4594.
doi:10.1210/endo.142.11.8547Kuhnz W, Blode H, Zimmermann H. Pharmacokinetics of exogenous natural and synthetic estrogens and antiestrogens. In: Oettel M, Schillinger E, eds. Estrogens and Antiestrogens II. Springer; 1999:261–322.
doi:10.1007/978-3-642-60107-1_13Canonico M, Oger E, Plu-Bureau G, et al; Estrogen and Thromboembolism Risk (ESTHER) Study Group. Hormone therapy and venous thromboembolism among postmenopausal women: impact of the route of estrogen administration and progestogens. Circulation. 2007;115(7):840–845.
doi:10.1161/CIRCULATIONAHA.106.642280Stanczyk FZ, Hapgood JP, Winer S, Mishell DR Jr. Progestogens used in postmenopausal hormone therapy: differences in their pharmacological properties, intracellular actions, and clinical effects. Endocr Rev. 2013;34(2):171–208.
doi:10.1210/er.2012-1008Tavaniotou A, Smitz J, Bourgain C, Devroey P. Comparison between different routes of progesterone administration as luteal phase and early pregnancy support. Hum Reprod Update. 2000;6(2):139–148.
doi:10.1093/humupd/6.2.139van den Heuvel MW, van Bragt AJ, Alnabawy AK, Kaptein MC. Comparison of ethinylestradiol pharmacokinetics in three hormonal contraceptive formulations: the vaginal ring, the transdermal patch and an oral contraceptive. Contraception. 2005;72(3):168–174.
doi:10.1016/j.contraception.2005.03.005Simmons KB, Haddad LB, Nanda K, Curtis KM. Drug interactions between non-rifamycin antibiotics and hormonal contraception: a systematic review. Am J Obstet Gynecol. 2018;218(1):88–97.
doi:10.1016/j.ajog.2017.07.003Reimers A, Brodtkorb E, Sabers A. Interactions between hormonal contraception and antiepileptic drugs: clinical and mechanistic considerations. Seizure. 2015;28:66–70.
doi:10.1016/j.seizure.2015.03.006