Systemic effects and endometrial safety of local estrogen therapy

Professor, Department of Obstetrics and Gynecology, New York University School of Medicine, Director, Gynecologic Ultrasound, Co-Director, Bone Densitometry, New York University Medical Center, New York, NY

DISCLOSURE
The author has received honoraria from Amgen, Boehringer Ingelheim, Eli Lily, Nordisk, Merck, and Pfizer. He is a consultant to Cook ObGyn and Philips Ultrasound; a speaker for Eli Lily and Warner Chilcott; and Director, Sonosite.

Vaginal atrophy is a prevalent condition resulting from estrogen deffciency that can occur at any time in a woman’s life but is more common after menopause.¹ The precise prevalence of vaginal atrophy is difficult to ascertain. A recent review of vaginal atrophy found reports ranging from 4% of women in early perimenopause to 47% of postmenopausal women. Fortunately, in most women, local or systemic estrogen preparations relieve vaginal atrophy symptoms.²

Interest in the potential for lower systemic hormonal exposure with use of local vaginal estrogen therapy has grown since the results of the Women’s Health Initiative (WHI) were made public almost 10 years ago.³ Vaginal administration of low-dose estrogen provides sufficient estrogen to relieve symptoms and reverse the atrophic changes associated with menopause, with the added benefit of limited systemic absorption.⁴ Therefore, local vaginal treatment is typically recommended for women seeking estrogen therapy solely for the treatment of vaginal atrophy.⁵

Estradiol tablets, estrogen creams, and estrogen rings are the three local estrogen formulations available in the United States. This review presents the clinical evidence about the safety of these formulations.

SYSTEMIC ABSORPTION OF VAGINAL ESTROGEN THERAPY

Absorption of estradiol and other steroid hormones is rapid, particularly through the thin vaginal epithelium characteristic of postmenopausal women.⁶ All three available formulations—the 17β-estradiol-releasing ring, 17β-estradiol tablets, and conjugated equine estrogen (CEE) and estradiol creams—are water-soluble and easily absorbed through the vaginal epithelium.⁷ Absorption with the ring or tablets may be slower than with creams.⁸ The concentration gradient across the epithelium determines the absorption rate; therefore, dosage may affect the extent of systemic exposure. Moreover, the distribution of absorbed estrogen within the pelvic region may differ depending on the anatomic position of administration. Some evidence suggests that delivery of estrogens to the lower one third of the vagina may limit distribution to the uterus, thereby lowering the risk of hyperplasia, and may be preferable for the treatment of vaginal atrophy.^8,9

An important diference between local vaginal and oral administration of estrogen is that vaginal administration circumvents metabolic and physiologic barriers to estrogen absorption.⁶ Consequently, substantially lower dosages of estrogen are required when applying estrogen directly to the vagina, compared with oral administration.¹⁰ Although circulating levels of estrogen do increase with local administration, serum estrogen levels typically remain relatively low.

ESTROGEN CREAM. Several small studies have examined elevations in plasma estrogen concentrations with administration of low-dose vaginal creams containing either CEEs or estradiol across a range of doses.^11,12 After 24 weeks of treatment at the highest dose of CEE cream used (1.25 mg/d), 21 of 59 (47%) women had a serum estradiol level outside the postmenopausal range (>49 pg/mL), although the magnitude of the elevation was not reported.¹² In an other study,¹¹ of 20 women treated for 6 months, CEE cream (0.3 g/d) produced only a minimal increase in estradiol and estrone levels.

Circulating levels of estradiol were measured in seven postmenopausal women treated with estradiol 10 µg vaginal cream for relief of symptoms of vaginal atrophy.¹³ After 3 weeks of daily administration, followed by an additional 9 weeks of twice-weekly treatment, circulating estradiol levels remained within the postmenopausal range.

No change in circulating mean estradiol (E2) levels have been reported with estriol cream, a vaginal estrogen therapy not available in the United States^14,15; according to a Cochrane Review, this is the only cream not associated with systemic absorption.¹⁶

Results from a Cochrane analysis of local estrogen therapy concluded that systemic absorption is greater with the CEE cream compared with the vaginal tablet.¹⁶ Because the administration of vaginal creams requires that patients measure and apply a dose themselves, the potential exists for them to use more estrogen than prescribed, increasing the likelihood of possible significant systemic absorption.⁴ Administration of estrogen with vaginal ring or tablets is not subject to the same potential for human error.

ESTROGEN RING. The estrogen ring formulation indicated for treatment of vaginal atrophy in the United States releases a consistent stable dose of approximately 7.5 µg of 17β-estradiol/24 h for 90 days.¹⁷ In an open-label study¹⁸ involving 129 women treated with a vaginal ring, no increase in serum estrogen levels were found after 3, 12, or 15 weeks of therapy. Similar results were found in 30 women aged ≥60 years using the ring for 6 months; no significant changes in serum estradiol or estrone levels were found.¹⁹ However, the same study showed that the use of the vaginal ring was associated with an increase in bone density, suggesting that some degree of systemic absorption occurred.

ESTROGEN TABLETS. Vaginal estradiol tablets (10-µg and 25-µg dosages) provide localized effects without appreciable absorption.^20,21 In a study²⁰ of 12 weeks of therapy with either 10 µg or 25 µg of estradiol, systemic absorption of estrogen was low and consistent over time. There was no evidence of accumulating levels of circulating estradiol. In a recently published comprehensive assessment²¹ of estradiol absorption following a dosage regimen of 10-µg and 25-µg estradiol vaginal tablets, postmenopausal women (60 to 70 years of age) with vaginal atrophy showed similar results. Findings revealed that systemic estrogen exposure is minimized with the ultra-low-dosage 10-µg estradiol vaginal tablet, compared with the 25-µg dose. As shown in FIGURE 1,²¹ following treatment with the 10-µg tablet, the E2 concentrations remained within the normal reference range for postmenopausal women; aside from a transient elevation on treatment day 1, values were low and similar to baseline (day –1). The profile of systemic exposure to E2 was similar with the 25-μg tablet, but absolute values were higher than with the lower dose (data not shown).

Because the 10-µg dose is adequate for optimum symptom relief with lower circulating E2 concentrations, the 25-µg tablet is no longer available (as of July 30, 2010).²²

ENDOMETRIAL EFFECTS

The primary concern regarding use of any estrogen therapy in women who have an intact uterus is the risk of endometrial hyperplasia and carcinoma associated with unopposed estrogen. Although available evidence suggests that low doses of locally administered vaginal estrogen are generally safe for the endometrium, data are limited. The Cochrane Review reported that of all the included studies, one case (2%) of hyperplasia was found among women using the vaginal ring and two cases (4%) in women using CEE cream.¹⁶ In a study²³ comparing CEE with tablets, endometrial proliferation occurred in seven cases (14%) among women using CEE cream and one woman (2%) using tablets. One woman in each group had endometrial proliferation in this study comparing cream and tablets over a period of 12 weeks. In the study¹² comparing the tablet versus CEE cream over 24 weeks, five (13%) women in the cream group and four (6%) in the tablet group had endometrial proliferation on biopsy.

The endometrial safety of the 10-µg estradiol vaginal tablet was most recently evaluated for the treatment of vaginal atrophy in 336 non hysterectomized postmenopausal women.²⁴ As shown in FIGURE 2,²⁴ baseline endometrial thick ness was 2.04 mm as determined by transvaginal ultrasonography (double layer) compared with 1.94 mm after 52 weeks of treatment. At study’s end, there was no evidence of increased endometrial proliferation or hyperplasia.

A 12-month study comparing the safety of the vaginal ring and the estradiol 25-µg vaginal tablet in 185 postmenopausal women identified significant endometrial thickening (>5 mm) in three cases in the ring group and in two cases in the tablet group.²⁵ In another clinical study evaluating the 10-µg estradiol dose, there was one case of endometrial adenocarcinoma stage II, which was considered unlikely to have developed in response to the relatively short course of study treatment.²⁶

Although endometrial hyperplasia has been seen with low-dose vaginal estrogen, it is rare, and the concomitant use of a progestin is not indicated.^4,27 A recent 6-month pilot study²⁸ of intravaginal administration of estriol 1 mg and progesterone 30 mg in 19 postmenopausal women found no cases of endometrial hyperplasia and, similar to results with unopposed estrogen, minimal systemic accumulation of estriol over the course of the study.

Overall, studies of endometrial effects of vaginally administered estrogen suggest that the incidence of hyperplasia or increased en-dometrial thickness is low and does not differ if treatment is administered as vaginal creams, rings, or tablets.¹⁶ Well-controlled studies with larger sample sizes need to be carried out to determine the benefit of using vaginal estriol and progesterone in postmenopausal women with vaginal atrophy.

CARDIOVASCULAR EFFECTS

Systemic estrogen use is known to increase the risk of thrombosis; the degree of risk depends on the dose and duration of therapy. Data evaluating the relationship of local estro-gen therapy with venous thromboembolism or other thrombotic events are limited. It is reasonable to assume that because systemic exposure is low with local estrogen therapy, the increase in the risk of thrombosis is low as well. Nonetheless, local estrogen therapy is contraindicated in women with active deep vein thrombosis, pulmonary embolism, arterial thromboembolic disease (for example, stroke and myocardial infarction), or a history of these conditions.^17,29,30

IMPACT ON BREAST CANCER

The estrogen plus progestin arm of the WHI³¹ was stopped prematurely because of the increased risk of invasive breast cancer observed in the active treatment arm. This finding with systemic estrogen plus progestin therapy in nonhysterectomized women raised concerns about the impact of even small elevations of circulating estrogen on breast cancer risk. Moreover, results from a meta-analysis of studies of systemic hormone therapy in women with a history of breast cancer concluded that the risk of recurrence is increased (relative risk, 3.41; 95% confidence interval, 1.59–7.33).³²

Regrettably, evidence on the impact of local vaginal estrogen therapy on breast cancer is limited.⁴ In a prospective observational study³³ of 18,314 women, vaginal estrogen therapy was not associated with an increased risk of breast cancer with no prior history.

Although systemic absorption of estro-gen with local estrogen therapy is likely to be minimal, it is unknown whether this limited absorption will affect outcomes in women with hormone-dependent cancers.⁴ Use of adjuvant therapy with aromatase inhibitors (AI) for estrogen-dependent breast cancer is associated with a high incidence of vaginal atrophy symptoms.³⁴ AIs act by blocking 95% of estrogen synthesis, typically resulting in circulating estradiol levels of <1 pg/mL.³⁵ As might be expected from studies in otherwise healthy women, vaginal administration of estradiol 25-µg tablets resulted in small increases in serum estradiol in a study³⁶ involving seven women receiving AI therapy. At day 14, the median serum estradiol level had increased from 0.82 pg/mL to 19.6 pg/mL. Although the increase is small and levels decreased to <10 pg/mL (median <5 pg/mL) by day 28, any rise above base line serum estradiol levels may have an impact on AI efficacy.

A preliminary study³⁵ on the use of local estrogen therapy in breast cancer survivors has been conducted. In this study, both low-dose vaginal estradiol tablet and estriol cream relieved vaginal atrophy, whereas a nonhormonal moisturizer provided only transient relief. Safety was evaluated by measuring serum estradiol levels and changes in endome-trial thickness, both of which were found to be minimal and clinically insignificant, although an evaluation of a longer duration of treatment is warranted. Because of the impact of moderate or severe vaginal atrophy on quality of life, women who do not respond to nonhormonal therapies may consider discussing the risks and benefits of local estrogen therapy.

TOLERABILITY

Vaginal estrogen use is generally well tolerated. In a comparative clinical trial,³⁷ with drawals due to adverse events were low (9 of 13 [7%] using the ring and 5 of 63 [8%] using CEE cream). This study found that 8 of 131 (6%) estradiol ring users and 5 of 63 (8%) CEE cream users experienced vaginal bleeding. In another 48-week trial,²⁵ 4 of 59 (7%) women randomized to use estradiol 25-µg tablets had vaginal bleeding. In this trial, none of the 126 women assigned to the estradiol ring group reported vaginal bleeding.

The incidence of breast tenderness, which is considered a marker of systemic exposure,¹⁶ is low for all three types of local estrogen therapy. However, evidence suggests that the incidence of breast tenderness may be higher with use of CEE cream. In the trial comparing use of the estradiol ring with CEE cream, breast tenderness contributed to discontinuation of study treatment in both groups.³⁷ In placebo-controlled trials of CEE cream, breast pain was reported in 8 of 143 (6%) women using a 12-week regimen of cream daily for 21 days followed by 7 days of no treatment. In the control group, breast pain occurred in 1 of 72 (1.4%) subjects.³⁰ Among those using a twice-weekly regimen, the incidence of breast pain was 3% in the active treatment group compared with none of 68 (0%) in the control group. The incidence of breast pain was <1% in pivotal trials of both the vaginal ring¹⁷ and estradiol 10-µg tablets.²⁹

PATIENT PREFERENCE

Because the three available vaginal estro-gen therapies do not show major differences in efficacy or safety, the choice of treatment may be determined by patient preference. In clinical trials, >80% of patients rate treatment with the estradiol ring or tablets as excellent to good. In the study comparing the ring to CEE cream, 84% of ring users rated treatment as excellent or good compared with 43% of cream users.^25,37

A retrospective analysis³⁸ of patient adherence to vaginal estrogen therapy in 13,074 women found that those prescribed vaginal creams discontinued treatment significantly sooner than those prescribed tablets (50% discontinuation at 30 days with cream and at 142 days with tablets; P < .001) and filled fewer prescriptions. Better adherence with the tablets occurred despite higher out-of-pocket costs for vaginal tablets.

SUMMARY AND CLINICAL IMPLICATIONS

• Vaginal estrogen therapy is safe and well tolerated in postmenopausal women with vaginal atrophy, although these conclusions are based on experience in small trials, largely of short duration.

• As with all forms of estrogen therapy, the Food and Drug Administration and NAMS have recommended the use of the lowest possible effective dose of vaginal estrogen for treating vaginal atrophy.^5,39

• Local estrogen treatments are associated with minimal systemic absorption.

• Endometrial hyperplasia with local vaginal therapy is rare. Use of progestin therapy is generally not needed in patients using low-dose local vaginal estrogen therapy.⁴

• In the absence of more rigorous studies suggesting otherwise, vaginal estrogens are not indicated for women receiving adjuvant AI therapy for breast cancer, and the potential for recurrence should be discussed with patients with a history of any hormone-sensitive cancer.

• Because efficacy and safety of the local estrogen formulations appear to be similar, the preparation preferred by the patient and the provider should be used. Therapy should be continued as long as the troublesome symptoms remain.

OBG Management ©2010 Quadrant Healthcom Inc.