Evidence on the use of progesterone in menopausal hormone therapy

Abstract

A need exists for a regulatory agency-approved hormone therapy (HT) with naturally occurring hormones combining progesterone (P4) and estradiol (E2), since no single product contains both endogenous hormones. Many women choose HT with P4 and millions of women around the world are using unapproved, poorly regulated compounded HT. The use of natural P4 in HT results, for the most part, in favorable outcomes without deleterious effects, as shown in clinical studies of postmenopausal women. Importantly, P4 used in HT prevents endometrial hyperplasia from estrogens while helping relieve vasomotor symptoms and improving quality-of-life measures. Additionally, risk of venous thromboembolism and breast cancer does not appear to increase with use of P4 plus estrogens as shown with synthetic progestins plus estrogens in large observations studies, and no detrimental effects of P4 in HT have been found on outcomes related to cardiovascular disease or cognition. A regulatory agency-approved HT with naturally occurring E2/P4 could be an option for the millions of women who desire a bioidentical product and/or are exposed to potential risks of inadequately studied and under-regulated compounded HT.

Keywords:

• Progesterone
• hormone therapy
• menopause

Introduction

The goal of hormone therapy (HT) is to relieve menopausal women of symptoms such as hot flushes, night sweats, dyspareunia, and vaginal dryness. When prescribing HT, either natural progesterone (P4) or a synthetic progestogen (progestin) is required for endometrial protection¹. Observational data suggest a safer profile of P4 versus progestins with respect to cardiovascular and breast cancer outcomes^2–6.

A high demand for HT containing P4 is shown by the estimated annual prescriptions of up to 21 million products containing natural P4, representing the most prescribed form of HT in the US⁷. Since no single FDA-approved formulation contains E2 and P4 combined, most of these (up to 18 million prescriptions) are for non-FDA-approved compounded hormone therapy⁷. Given that compounded HT is not adequately studied and under-regulated, such high exposure of menopausal women to compounded HT is concerning since certain risks may be associated with compounded HT. Recent evidence suggests an increased risk of endometrial hyperplasia and cancer in users of compounded HT^8–10.

A regulatory agency-approved menopausal HT with E2 and P4 could help fill the need for women to be safely treated with a natural or bioidentical product for their bothersome vasomotor symptoms (VMS)^11,12, shown to negatively impact their quality of life, sleep, and work productivity^11,13. This paper reviews historical and more recent clinical data supporting the use of P4, by various routes of administration, in combination with estrogens (E2 or conjugated equine estrogens (CEE)) as menopausal HT.

Methodology

PubMed was searched since its inception for clinical studies on HT containing natural P4. Relevant studies, including randomized, controlled trials of estrogens with P4, and large observational studies comparing different progestins with P4 were the focus of this article. Studies on transdermal P4 were not included since no adequately designed trials have shown it can deliver adequate P4 for endometrial protection¹⁴.

Keywords such as bone, breast, breast cancer, cardiovascular (CVD), cognition, endometrium, hot flush, lipids, mammographic breast density, myocardial infarction, quality of life, sleep, stroke, vasomotor symptoms, and venous thromboembolism (VTE) were used in conjunction with P4 and menopause. Limits were the English language, keywords within the title or abstract, clinical trials, and review articles. Bibliographies of relevant review articles were also reviewed.

Endometrial safety

Progesterone and progestins have a pronounced effect on endometrial morphology, with different potencies and dosages inducing qualitative and quantitative morphologic changes. Randomized, controlled trials show progestogens at the right dosage prevent endometrial hyperplasia^15,16 and endometrial cancer¹⁷; however, oral P4 has not been thought to protect the endometrium as well as progestins, based on European observational data (almost exclusively from France)^18,19. Although an increased risk of endometrial cancer was found with current use^18,19 or >5 years use¹⁹ of estrogens plus P4, but not with use of estrogens plus synthetic progestins, in the Etude Epidémiologique auprès de femmes de l’Education Nationale (E3N) and European Prospective Investigation into Cancer and Nutrition (EPIC) trials, definitive conclusions are difficult to make from these reports since risk was not analyzed by dose or timing (i.e. cyclic vs. continuous) of progestogen use.

Effectiveness of P4 in protecting the endometrium was first supported by proof-of-concept studies showing a dose-dependent inhibition of endometrial proliferation in women taking various doses of cyclic P4 with estrogens^20,21. In the Postmenopausal Estrogen/Progestin Interventions (PEPI) trial (n = 118–120 in each group), cyclic 200 mg P4 (12 days) plus 0.625 mg CEE provided endometrial hyperplasia rates similar to those of placebo¹⁵. Based on these data¹⁵, the only FDA-approved P4 regimen for endometrial protection is cyclic 200 mg with CEE²². However, the PEPI study was not powered adequately to detect endometrial hyperplasia or cancer risk. To demonstrate adequate endometrial protection (≤1% with <4% upper bound of 95% confidence interval), regulatory agencies world-wide currently recommend a 1-year clinical trial with at least 300 subjects^23,24.

A preliminary report of the phase-3 REPLENISH trial showed multiple doses of the investigational, oral, continuous combined, solubilized E2/P4 (TX-001HR, TherapeuticsMD, Boca Raton, FL, USA: 1 mg/100 mg, 0.5 mg/100 mg, 0.5 mg/50 mg, 0.25 mg/50 mg) resulting in no cases of endometrial hyperplasia or cancer (n = 274–306 per group) after 12 months²⁵. This was the first sufficiently powered trial to identify oral P4 doses that antagonized the endometrial effects of different E2 doses (1 mg, 0.5 mg, or 0.25 mg)²⁵.

Vaginal P4 has also been given for endometrial protection, as it has been hypothesized to have better access to endometrial tissues through the ‘first uterine pass effect’ and possibly fewer adverse effects^26,27. While a significantly higher endometrial-to-serum P4 ratio with vaginal P4 compared with intramuscular P4 injection has been shown in a small study²⁷, no studies large enough adequately to evaluate endometrial protection with vaginal P4 have been conducted. The Early versus Late Intervention Trial with Estradiol (ELITE), the only 5-year, randomized trial comparing 1 mg/day oral E2 or placebo with 45 mg/day of cyclical vaginal P4 gel (10 days/cycle)²⁸, found no significant difference in E2/P4 vs. placebo for endometrial cancer (assessed as an adverse event; reported by Stute and colleagues¹⁴). Some smaller studies have shown vaginal P4 to prevent endometrial hyperplasia when given cyclically or continuously as a gel with oral or transdermal estrogens^29–31, or as a vaginal ring³², but not as a vaginal P4 cream³³.

In conclusion, the REPLENISH trial is the first and only adequately designed trial showing oral, continuous P4 to prevent endometrial hyperplasia when given at the right ratio.

Uterine bleeding

Uterine bleeding has been related to an imbalance between angiogenic and antiangiogenic factors in the endometrial vasculature³⁴. Such alterations could induce proangiogenic activity in the endometrial capillary plexus, with consequent aberrant vasculogenesis that may result in irregular endometrial bleeding³⁴. In vitro studies have shown that the angiogenic imbalance is less pronounced with P4 than with synthetic progestogens^35,36. In addition, clinical studies demonstrated that the ratio of estrogen to progestogen may play a role in determining the risk of vaginal bleeding³⁷. However, few head-to-head studies have evaluated the bleeding profiles of various HT regimens of estrogens with progesterone compared with HT containing other progestogens.

In a 9-month study of women taking CEE randomized to either micronized 200 mg/day P4 (n = 89) or 5 mg/day medroxyprogesterone acetate (MPA; n = 93), more women taking P4 versus MPA had amenorrhea (p = 0.001), and women taking P4 had fewer bleeding days (p = 0.01) and less blood flow (p < 0.001)³⁸. Significantly less vaginal bleeding was seen with oral P4 than MPA (p < 0.01) in a small study (n = 23) of alternating 2-week cycles of CEE alone (0.625 mg) or CEE plus progestogen (cyclic MPA 5 mg/day or P4 200 mg/day)³⁹. P4 also achieved a high rate of amenorrhea (91.6%) in a study of 98 women treated for 6 months with transdermal E2 1.5 mg/day plus oral P4 100 mg/day on days 1–21 or 25⁴⁰. Preliminary data from the REPLENISH trial showed that several continuous combined doses of E2/P4 (TX-001HR) resulted in high rates of amenorrhea that improved over time (56–73% from cycles 1–13)⁴¹. In contrast, Di Carlo and colleagues⁴² reported that sequential HT containing progesterone had more episodes of irregular bleeding than with sequential HT containing synthetic progestins. The same investigators reported that users of HT with sequential vaginal progesterone had more regular bleeding episodes, fewer irregular bleeding episodes and more amenorrhea than users of HT with sequential oral progesterone⁴³.

Menopausal symptoms

Although progestogens are included in HT to provide endometrial protection, some progestogens, including P4 at high dosage, have been shown to provide relief of VMS when given without an estrogen. A 12-week, randomized study in postmenopausal women (n = 133) found that daily VMS frequency improved significantly more with oral 300 mg/day P4 alone than with placebo⁴⁴.

Data from the REPLENISH trial show that, for the most part, the frequency of moderate-to-severe VMS significantly improved from week 3 to 12 for E2/P4 doses of 1 mg/100 mg, 0.5 mg/100 mg, and 0.25 mg/50 mg, and at weeks 6–12 for 0.5 mg/50 mg compared with placebo (p < 0.05)²⁵. Significant reductions in VMS severity were observed at weeks 3–12 for 1 mg E2/100 mg P4 and 0.5 mg E2/100 mg P4, and at most weeks between 6 and 12 for 0.5 mg E2/50 mg P4 and 0.25 mg E2/50 mg P4²⁵.

Head-to-head studies, including the PEPI trial, show similar relief of hot flushes with CEE plus cyclic or continuous P4 or MPA^45–47.

Quality of life

One study suggests positive effects on quality-of-life outcomes with HT regimens containing 0.625 mg CEE with MPA 5 mg/day or micronized P4 200 mg/day (n = 182) for 9 months³⁸. In the REPLENISH trial, E2/P4 (1 mg/100 mg, 0.5 mg/100 mg, 0.5 mg/50 mg) significantly improved quality of life as measured by MENQOL at 12 weeks and 12 months⁴⁸.

Sleep

Progesterone is converted to allopregnanolone in the mammalian brain, which has been shown to induce GABAergic effects and promote sleep. Sleep parameters improved in several trials of P4-containing HT, with some showing better improvements when P4 vs. MPA was used. A polysomnography study of 21 menopausal women found that 0.625 mg CEE with cyclic 200 mg/day P4, but not with cyclic 5 mg/day MPA, significantly improved sleep efficiency and decreased time spent awake after sleep onset from baseline, with similar improvements in other subjective sleep indices in both groups⁴⁶. In another small, cross-over, polysomnography study (n = 10), 300 mg P4 alone for 21 days significantly reduced time spent awake and increased rapid eye movement sleep in the first third of the night vs. placebo⁴⁹. Continuous CEE 0.3 mg/day with P4 100 mg also provided significantly better sleep outcomes than CEE/MPA 2.5 mg⁴⁷. The REPLENISH trial showed improved sleep parameters (Medical Outcomes Study – Sleep Scale) with various continuous E2/P4 doses at week 12 and months 6 and 12⁵⁰, consistent with a reduction in hot flushes.

Bone

Effects of HT containing P4 or MPA appear to be similar on bone mineral density (BMD). This was shown for the spine and hip in the PEPI trial with CEE plus cyclic MPA or cyclic P4 vs. placebo at 3 years⁵¹, and in a smaller, 3-year, randomized, open-label study (n = 60) of transdermal E2 gel 0.75 or 1.5 mg/day with P4 100 mg/day or MPA 2 mg/day (25 days per month)⁵². Other studies show BMD improvements with estrogens plus cyclic or intermittent P4 use^53–55. In a KEEPS substudy (n = 76), oral CEE or transdermal E2 patch with cyclic oral micronized P4 for 4 years prevented the cortical volumetric BMD decline and cortical porosity increase at the distal radius seen with placebo⁵³.

Breast

Effects of HT on breast cell proliferation in premenopausal and postmenopausal women have been evaluated. Premenopausal⁵⁶ and postmenopausal⁵⁷ women scheduled for breast surgery, for presumably benign lesions, were randomized to transdermal gel containing 1.5 mg E2, 25 mg P4, combined E2/P4, or placebo applied daily to the breast for approximately 14 days before surgery. For the most part, in both studies, proliferation markers (mitotic index and proliferating cell nuclear antigen index) increased with E2 vs. placebo, decreased with P4 vs. placebo, and were comparable between E2/P4 and P4 or placebo^56,57. In another study of postmenopausal women (n = 77) taking two cycles of 0.625 mg CEE plus cyclic 5 mg MPA or 1.5 mg E2 gel plus cyclic oral 200 mg P4, breast cell proliferation significantly increased from baseline with CEE/MPA, but not with E2/P4^58,59.

New-onset breast tenderness after initiation of HT containing a synthetic progestin has been associated with increased mammographic density and breast cancer risk^60,61. A KEEPS substudy (n = 116) that evaluated breast pain over 4 years reported no significant increases in pain over time and no differences in ‘worst’ pain or ‘average’ pain at any time point with oral 0.45 mg/day CEE or transdermal 50 μg/day E2 (both given with cyclic 200 mg/day P4) or placebo⁶². In a small study (n = 23), women were given placebo, then cyclic 5 mg/day MPA or 200 mg/day P4 alone followed by 2 months of 0.625 mg/day CEE with progestogen in the last 2 weeks/month³⁹. While neither progestogen was associated with breast tenderness when given without CEE, breast tenderness was greater with CEE/MPA than with CEE/P4³⁹.

When mammographic breast density was evaluated in the PEPI trial, women randomized to CEE alone, CEE/cyclic MPA, CEE/continuous MPA, or CEE/cyclic P4 had 0.9%, 4.6%, 4.4%, and 3.1% increases in breast density, respectively, compared with placebo (−0.4%; p < 0.001 for combined progestogen group vs. placebo, with no difference among progestogen groups)⁶³. A smaller Swedish study showed 18.9% of patients taking oral CEE 0.625 mg plus cyclic MPA 5 mg (n = 37) had an increase in BI-RADS category of at least one level from baseline (p = 0.01 vs. baseline) compared with 6.3% of women taking percutaneous E2 gel 1.5 mg plus oral P4 200 mg (n = 32; p = NS)⁵⁹.

Large observational studies have compared effects of HT containing progestins or P4 on breast cancer risk. In two publications from European studies (n = 54 548 (E3N cohort) and 80 377 (E3N-EPIC cohort)), an increase in breast cancer risk was found in women who currently took estrogen plus synthetic progestogens but not estrogens plus P4 (risk ratios reported in Table 1; one study⁶⁴ combined P4 with dydrogesterone)^64–66. Breast cancer risk by duration of HT use is summarized in Table 2. In the most recent analysis of E3N, increased risk of breast cancer was found with current use of estrogen plus P4/dydrogesterone for >5 years, and current use of estrogen plus synthetic progestogens for ≤5 years and >5 years with a greater magnitude than with P4⁶⁴. In addition, current use of estrogen plus P4/dydrogesterone for ≤5 or >5 years in women with a body mass index (BMI) < 25 kg/m², or estrogen plus synthetic progestogens for ≤5 or >5 years independent of BMI, had an increased breast cancer risk, but with a greater magnitude in women taking synthetic progestins⁶⁴. Since P4 and dydrogesterone were combined in the analysis, making definitive conclusions about P4 is not possible⁶⁴. Finally, while risk with estrogens plus P4 disappeared in past users, it remained in users of estrogens plus synthetic progestins⁶⁴. In the initial report of the E3N cohort⁶⁶, no significantly increased risk was found at 2-year increments for estrogens plus progesterone or dydrogesterone, but increased risk was consistently seen with estrogens plus synthetic progestins over the same time periods. Analysis by duration of use in the initial combined E3N–EPIC cohort found an increase in risk with transdermal estrogens plus synthetic progestins vs. P4 for the first 4 years of use⁶⁵. While an increased risk was shown with 2–4 years or >4 years of oral estrogens plus synthetic progestin use, risk with oral estrogens plus P4 was not analyzed due to the small number of women taking it⁶⁵.

Table 1. Risk of breast cancer and venous thromboembolism with hormone therapy containing progesterone or progestins in large observational studies.

Study	Estrogen and/or progestogen	Hazard ratio (95% confidence interval)
Breast cancer
Fournier et al., 2005^65	Estrogen alone	1.1 (0.8–1.6)
	Estrogen plus progesterone	0.9 (0.7–1.2)
	transdermal estrogen	0.9 (0.7–1.2)
	oral estrogen	No events
	Estrogen plus synthetic progestins	1.4 (1.2–1.7)
	transdermal estrogen	1.4 (1.2–1.7)
	oral estrogen	1.5 (1.1–1.9)
Fournier et al., 2008^66	Oral estrogen alone	1.32 (0.76–2.29)
	Oral estrogen plus progestogen
	progesterone	Not analyzed^a
	dydrogesterone	0.77 (0.36–1.62)
	medrogestone	2.74 (1.42–5.29)
	chlormadinone acetate	2.02 (1.00–4.06)
	cyproterone acetate	2.57 (1.81–3.65)
	promegestone	1.62 (0.94–2.82)
	nomegestrol acetate	1.10 (0.55–2.21)
	norethisterone acetate	2.11 (1.56–2.86)
	medroxyprogesterone acetate	1.48 (1.02–2.16)
	Transdermal estrogen alone	1.28 (0.98–1.69)
	Transdermal estrogen plus progestogen
	progesterone	1.08 (0.89–1.31)
	dydrogesterone	1.18 (0.95–1.48)
	medrogestone	2.03 (1.39–2.97)
	chlormadinone acetate	1.48 (1.05–2.09)
	cyproterone acetate	Not analyzed^a
	promegestone	1.52 (1.19–1.96)
	nomegestrol acetate	1.60 (1.28–2.01)
	norethisterone acetate	Not analyzed^a
	medroxyprogesterone acetate	Not analyzed^a
Fournier et al., 2014^64	Estrogen alone	1.17 (0.99–1.38)
	Estrogen plus progesterone/dydrogesterone	1.22 (1.11–1.35)
	Estrogen plus synthetic progestins	1.87 (1.71–2.04)
Cordina-Duverger et al., 2013^55	Estrogen alone (any)	1.19 (0.69–2.04)
	Estrogen (any) with progestogens	1.33 (0.92–1.92)
	natural progesterone	0.80 (0.44–1.43)
	synthetic progestins	1.72 (1.11–2.65)
	progesterone derivatives	1.57 (0.99–2.49)
	testosterone derivatives	3.35 (1.07–10.4)
Venous thromboembolism
Canonico et al., 2010^3	Oral estrogens	1.7 (1.1–2.8)
	Transdermal estrogens	1.1 (0.8–1.8)
	Any estrogen with progestogens
	progesterone	0.9 (0.6–1.5)
	pregnane derivatives	1.3 (0.9–2.0)
	nortestosterone derivatives	1.4 (0.7–2.4)
	norpregnane derivatives	1.8 (1.2–2.7)
Canonico et al., 2007^4	Oral estrogens	4.2 (1.5–11.6)
	Transdermal estrogens	0.9 (0.4–2.1)
	Any estrogen with progestogens
	progesterone	0.7 (0.3–1.9)
	pregnane derivatives	0.9 (0.4–2.3)
	norpregnane derivatives	3.9 (1.5–10.0)
Olie et al., 2011 (recurrent)^5	Oral estrogens	6.4 (1.5–27.3)
	Transdermal estrogens ± progestogens	1.0 (0.4–2.4)
	estrogen alone	1.1 (0.2–8.1)
	progesterone	1.0 (0.3–3.2)
	Norpregnane derivatives	4.7 (1.1–20.0)
	pregnane derivatives	No events
Ischemic stroke
Canonico et al., 2016^6	Oral estradiol	1.58 (1.01–2.49)
	Transdermal estradiol	0.83 (0.56–1.24)
	Estradiol with progestogens
	progesterone	0.78 (0.49–1.26)
	pregnane derivatives	1.00 (0.60–1.66)
	nortestosterone derivatives	1.26 (0.62–2.58)
	norpregnane derivatives	2.25 (1.05–4.81)

^a Not analyzed, less than five cases of breast cancer.

Table 2. Risk of breast cancer with hormone therapy containing progesterone or progestins by duration of use in large observational studies.

Study	Estrogen and/or progestogen	Hazard ratio (95% confidence interval)
Fournier et al., 2005^65	Transdermal estrogens plus progesterone
	<2 years	0.9 (0.6–1.4)
	2–4 years	0.7 (0.4–1.2)
	≥4 years	1.2 (0.7–2.0)
	Transdermal estrogens plus synthetic progestins
	<2 years	1.6 (1.3–2.0)
	2–4 years	1.4 (1.0–1.8)
	≥4 years	1.2 (0.8–1.7)
	Oral estrogens plus synthetic progestins^a
	<2 years	1.2 (0.9–1.8)
	2–4 years	1.6 (1.1–2.3)
	≥4 years	1.9 (1.2–3.2)
Fournier et al., 2008^66	Estrogens plus progesterone
	<2 years	0.71 (0.44–1.14)
	2–4 years	0.95 (0.67–1.36)
	4–6 years	1.26 (0.87–1.82)
	≥6 years	1.22 (0.89–1.67)
	Estrogens plus dydrogesterone
	<2 years	0.84 (0.51–1.38)
	2–4 years	1.16 (0.79–1.71)
	4–6 years	1.28 (0.83–1.99)
	≥6 years	1.32 (0.93–1.86)
	Estrogens plus other progestogens
	<2 years	1.36 (1.07–1.72)
	2–4 years	1.59 (1.30–1.94)
	4–6 years	1.79 (1.44–2.23)
	≥6 years	1.95 (1.62–2.35)
Fournier et al., 2014^64	Estrogens plus progesterone/dydrogesterone
	≤5 years	1.13 (0.99–1.29)
	>5 years	1.31 (1.15–1.48)
	Estrogens plus other progestogens
	≤5 years	1.70 (1.50–1.91)
	>5 years	2.02 (1.81–2.26)

^a Oral estrogens with progesterone not analyzed due to low number of women who took it.

In a French case–control study (n = 1232), current use of estrogen plus P4 did not increase breast cancer risk, while estrogen plus synthetic progestogens significantly elevated the risk, particularly testosterone derivatives (Table 1)². A systematic review and meta-analysis of studies (n = 14 studies) of women using E2 with or without progestogen showed no association between E2 only and risk of breast cancer; however, risk was significantly elevated when E2 was combined with MPA, norethisterone acetate, or levonorgestrel, but not with P4 or dydrogesterone⁶⁷.

Cardiovascular disease

Venous thromboembolism

Micronized P4 does not appear to increase the risk of VTE in postmenopausal women. Both the French case–control ESTHER (Estrogen and Thromboembolism Risk) and French E3N cohort studies reported that concomitant use of P4 or pregnane derivatives with estrogens did not increase the VTE risk in postmenopausal women, while concomitant use of norpregnane derivatives did (Table 1)^3,4. Similarly, the Menopause, Estrogen, and Veins cohort study suggested no additional risk of P4 on recurrent VTE in women taking transdermal estrogens, whereas norpregnane derivatives with transdermal estrogens were associated with significantly increased risk (Table 1)⁵. In the French Study of NorpregnAnes on Coagulation, concentrations of activated protein C (APC) sensitivity and prothrombin fragment 1 + 2 were similar with transdermal estrogens plus P4 vs. non-users, while women who used concomitant norpregnane derivatives were less sensitive to APC and had higher prothrombin fragment 1 + 2 than non-users⁶⁸.

Cardiovascular disease outcomes

Only two randomized, controlled studies examined the effects of HT with P4 on cardiovascular outcomes. ELITE tested the hormone-timing hypothesis, that early but not later initiation of HT would prevent subclinical atherosclerosis progression²⁸. Healthy postmenopausal women (n = 643) stratified by time since menopause (<6 years or ≥10 years) were randomized to 1 mg/day oral E2 or placebo; women with a uterus received 45 mg/day vaginal P4 gel or placebo applied cyclically for 10 days/cycle²⁸. At a median 5-year follow-up, carotid artery intima-media thickness (CIMT) progression was significantly lower with E2 than with placebo only in early postmenopause (p = 0.008) but not later in menopause (p = 0.29), and the effects of E2 alone and E2/P4 were similar²⁸. In KEEPS, women (n = 727) in early menopause (within 6 and 36 months) were randomized to continuous oral CEE 0.45 mg/day or transdermal E2 50 μg/day patch (plus oral P4 200 mg/day for 12 days/month)⁶⁹. Vascular imaging showed comparable increases in both CIMT and coronary artery calcium scores over 4-year follow-up between groups⁶⁹.

Progesterone also appears to be beneficial in women who have myocardial ischemia. In a cross-over trial of women with stable exertional angina, myocardial ischemia, coronary artery disease and/or previous myocardial infarction using a treadmill, 4 weeks of E2 (2 mg/day) improved their exercise time to myocardial ischemia, which was further improved after receiving 12 days transvaginal P4 gel (90 mg every other day) but not oral MPA (10 mg/day)⁷⁰.

Lastly, a French case–control study found an increased risk of ischemic stroke in women who used oral but not transdermal estradiol, and an increased risk with concomitant use of norpregnane derivatives, but not P4, pregnane derivatives, or nortestosterone derivatives (Table 1)⁶.

Lipids

Well-established effects of oral estrogens on lipids are increases in high density lipoprotein (HDL) cholesterol decreases in total cholesterol and low density lipoprotein (LDL) cholesterol, and increase in triglycerides, while these effects are not apparent with transdermal estrogens. Addition of P4 has been shown to enhance or have neutral effects on the beneficial changes of oral estrogens on lipids.

In the randomized, controlled PEPI trial, all active treatment groups increased HDL cholesterol relative to a decline in the placebo group after 3 years, and HDL cholesterol improvements were significantly greater in the CEE and CEE/P4 groups vs. the CEE/MPA groups⁷¹. All active regimens decreased LDL cholesterol and increased triglycerides vs. placebo⁷¹. Similarly in another study, 0.3 mg CEE plus P4 100 mg/day resulted in higher HDL cholesterol and triglycerides, lower LDL cholesterol and non-HDL cholesterol, and lower total cholesterol/HDL cholesterol ratio compared with baseline (all, p < 0.05)⁷². In a study of transdermal E2 gel alone or with cyclic oral P4 (200 mg), or 0.625 mg CEE alone or with cyclic oral P4, adding P4 to CEE enhanced the elevated HDL cholesterol level, preserved the decrease in LDL cholesterol, and lowered the elevation in triglycerides, while P4 added to the transdermal E2 regimen prevented an increase in HDL cholesterol⁷³.

In other studies, neutral or beneficial effects of P4 were observed. In an open-label, randomized trial (n = 107) of 12 cycles, CEE/dydrogesterone was associated with a significant increase in triglycerides, while CEE/P4 had a neutral effect; both therapies significantly elevated HDL cholesterol and lowered LDL cholesterol levels⁷⁴. Two other randomized, cross-over trials showed a neutral or beneficial effect of P4 in non-oral estrogen regimens in early postmenopausal women^75,76 and a transient decrease in HDL cholesterol was seen with a levonorgestrel-containing intrauterine device, but not with oral or vaginal P4⁷⁷.

Endothelial function

Several trials have confirmed that P4 has no adverse vascular effects. In a KEEPS substudy, recently postmenopausal women without CVD at baseline who received oral CEE 0.45 mg/day or transdermal E2 50 μg/day combined with cyclic oral P4 (200 mg/day) had no significant change in reactive hyperemia index (used to assess endothelial function) with treatment⁷⁸. A study of forearm blood flow in response to brachial artery infusions of vasodilators showed that administration of P4 either alone or with E2 had no detrimental effects in postmenopausal women without cardiovascular risk factors⁷⁹. In a 12-week, cross-over study of women receiving oral CEE (0.625 mg) or transdermal E2 (50 μg) combined with continuous oral P4 (100 mg), CEE/P4 significantly decreased E-selectin, P-selectin, and intercellular adhesion molecule (ICAM-1) and increased C-reactive protein, while transdermal E2/P4 only decreased ICAM-1 levels⁸⁰. Furthermore, endothelial function was improved and inflammation markers decreased in recently menopausal women who received 3 months of transdermal E2 (50 μg) plus cyclic oral P4 (200 mg) compared with non-users⁸¹.

Expression of vascular cell adhesion molecule (VCAM-1) has been shown to be one of the earliest events that occur in the atherogenic process⁸². In human umbilical vein endothelial cells, pretreatment with P4, but not MPA, inhibited TNFα-induced increases in VCAM-1 protein expression and mRNA accumulation⁸². Similarly, significant reductions in IL-1β-stimulated VCAM-1 and ICAM-1 expression were observed when female human iliac artery endothelial cells were pretreated with E2/P4 (but not either alone)⁸³.

Cognition

Progesterone combined with estrogens does not appear to have any deleterious or beneficial effects on cognition in postmenopausal women, as shown by the limited available evidence. The timing effect of oral E2 (1 mg) vs. placebo on cognition was evaluated in the ELITE-Cog trial (n = 567); women with a uterus also received 4% vaginal P4 gel (45 mg) or placebo 10 days/cycle⁸⁴. Compared with placebo, E2/P4 had no negative effects on verbal memory, executive functions, or global cognition after 2.5 years of treatment in women who initiated treatment within 6 years or ≥10 years after menopause⁸⁴. Similarly, the 4-year KEEPS-Cognitive and Affective Study found no significant difference in cognitive variables between recently menopausal women who received oral CEE (0.45 mg) or transdermal E2 (50 μg), both combined with cyclic oral P4 (200 mg), or placebo⁸⁵. In a small, 12-week trial of recently menopausal women who received 0.625 mg CEE plus cyclic 10 mg MPA (n = 9), 200 mg P4 (n = 8), or placebo (n = 7), only CEE/P4 provided significant improvements from baseline in delayed verbal memory and working memory tests⁸⁶.

Summary

A need exists for regulatory agency-approved HT with naturally occurring hormones such as P4 and E2, as use of compounded HT is on the rise, exposing millions of women to potential risk. Existing evidence from clinical studies on the use of natural P4 in HT, for the most part, shows favorable outcomes, without deleterious effects. Both E2 and CEE combined with P4 relieve vasomotor symptoms and improve measures of quality of life. Additionally, HT with P4 has been shown to prevent endometrial hyperplasia from estrogens, and large observational studies show risk of VTE and breast cancer does not increase with use of P4 with estrogens. Further, no detrimental effects of P4 on outcomes related to cardiovascular disease or cognition have been reported. A recent randomized, controlled study, REPLENISH, showed that an oral, combined E2/P4 capsule at various doses improved hot flushes without endometrial hyperplasia in menopausal women. A regulatory agency-approved HT with naturally occurring P4 and E2 could be an option for the millions of women exposed to potential risks with inadequately studied compounded HT.

Conflict of interest

Dr Mirkin is an employee of TherapeuticsMD.

Acknowledgements

The medical writing assistance provided by Kathleen Ohleth, PhD, CMPP and Dominique Verlaan, PhD, CMPP of Precise Publications, LLC is greatly appreciated.

Additional information

Funding

Medical writing assistance provided by Kathleen Ohleth, PhD, CMPP and Dominique Verlaan, PhD, CMPP of Precise Publications, LLC was supported by TherapeuticsMD.