Estrogen therapy remains the only treatment approved by the U.S. Food and Drug Administration (FDA) for menopausal women with vasomotor symptoms and vulvovaginal atrophy.1–4 The need for such therapy persists despite data from the Women’s Health Initiative (WHI) that describe long-term risks in a population of users. Initial analyses demonstrated an increased risk of breast cancer, stroke, coronary heart disease (CHD), and dementia in a population of postmenopausal women receiving daily, combined estrogen and progestin treatment5,6 and an increased risk of strokes and dementia, but not cardiovascular events or breast cancer, in women receiving estrogen only.7,8 More recent analyses found that compared with placebo treatment, the rate of cardiovascular events may be decreased in women aged 50–59 years receiving estrogen only.9 Although an understanding of the nature of the risks of estrogen therapy continues to evolve, most experts agree, and the FDA now requires in its “class labeling,” that the lowest possible dose of estrogen should be prescribed to treat postmenopausal women with vasomotor symptoms for the shortest time consistent with therapeutic goals.10–13 This dose should bring clinical relief to patients at the lowest possible serum estradiol (E2) concentration to minimize estrogen-related adverse effects and potentially limit the rare, but serious, long-term risks of exogenous estrogen treatment.14 The objective of this study was to estimate the efficacy and safety of a new transdermal gel containing 0.06% E2 (Elestrin, BioSante Pharmaceuticals, Inc., Lincolnshire, IL), aiming to identify the lowest effective dose for the treatment of vasomotor and vulvovaginal atrophy symptoms in postmenopausal women in accord with current FDA guidance.15
MATERIALS AND METHODS
This phase III, randomized, double-blind, placebo-controlled, multicenter, parallel-group study was conducted at 28 sites in the United States and two sites in Canada. The protocol was approved by review boards at each institution, and the study was conducted in accordance with the Declaration of Helsinki (1996 revision). Written informed consent was obtained from all study participants.
Participants were healthy women aged 18 years or older who had undergone natural (amenorrhea for 12 months or more) or surgical menopause (bilateral oophorectomy with or without hysterectomy 6 months or more before screening), who had serum E2 less than 20 pg/mL, follicle stimulating hormone (FSH) more than 40 milli-international units/mL, and body mass index of 18 to 35 kg/m2. Women were excluded if they had a history of estrogen-dependent neoplasia; endometrial hyperplasia; active hepatic, gallbladder, renal, or endocrine disease other than controlled thyroid abnormalities; or if they were receiving concomitant medications that could potentially interfere with hot flush frequency, severity, or their assessment. Eligible participants recorded the number, time, and severity (0, none; 1, mild; 2, moderate; or 3, severe) of hot flushes over the first 14 days of a 3- to 4-week screening period. Participants who experienced 60 or more moderate-to-severe hot flushes each week and were otherwise eligible had blood taken to measure serum E2 and entered a 1-week single-blind placebo lead-in period, during which they applied placebo gel once daily. Participants who successfully completed the placebo lead-in period by finishing the diary appropriately and had a serum E2 less than 20 pg/mL, then completed a vaginal health self-assessment and were randomly assigned to groups and entered the 12-week double-blind treatment period. Participants, investigators, and outcome assessors were blinded to treatment assignment, and no unblinding occurred during the trial. To maintain the study blind, participants applied the study drug from two metered-dose bottles (0.87 g gel per pump actuation), one labeled Bottle A and the other Bottle B, that contained placebo gel or E2 gel as necessary to deliver the appropriate treatment and dose.
Study coordinators assigned treatment numbers to participants entering the 12-week double-blind treatment period. The participant randomization number was printed on the label of each participant drug kit, and on each drug bottle within the kit. Initially, participant numbers were randomly allocated in blocks of three and in a 1:1:1 ratio to E2 gel 1.7 g/d (1.02 mg E2), E2 gel 2.6 g/d (1.56 mg E2) or placebo gel using a computer-generated randomization list. The study was amended after discussion with the FDA in light of WHI results5–8 to identify the lowest effective dose of E2 gel15 and limit the number of participants exposed to the 2.6 g/d dose. After the amendment, the E2 gel 0.87 g/d dose (0.52 mg E2) was added and the 2.6 g/d dose was discontinued from further enrollment. Participant numbers were then assigned in a 4:2:2 ratio to 0.87 g/d E2 gel, 1.7 g/d E2 gel, or placebo gel using a computer-generated randomization list. At study completion, four treatment groups were available for evaluation: E2 gel 0.87 g/d (n=136), E2 gel 1.7 g/d (n=142), E2 gel 2.6 g/d (n=69), and placebo (n=137) (Fig. 1).
During the 12-week double-blind treatment period, participants applied gel once daily to a small area of the upper arm and recorded in diaries the daily number and severity of hot flushes and weekly self-assessments of the severity of five vulvovaginal atrophy symptoms (vaginal dryness, vaginal irritation or itching, pain or difficulty passing urine, vaginal pain with sexual activity, and vaginal bleeding with sexual activity). Vaginal atrophy severity was assessed as “none” or “no sexual activity”; “a little [mild]”; “quite a bit [moderate]”; or “extreme [severe],” respectively scored as 0, 1, 2, or 3. Vulvovaginal atrophy was also assessed by vaginal pH, vaginal maturation index (calculated by 0.5×[100×I/(P+I+S)] + [100×S/(P+I+S)], where I is the number of intermediate cells, P is the number of parabasal cells, and S is the number of superficial cells in vaginal wall samples), and physician assessment of vaginal atrophy during gynecologic examinations at screening and end of treatment.
The coprimary efficacy endpoints for vasomotor symptoms were mean change from baseline to weeks 4 and 12 in moderate-to-severe hot flush rate and hot flush severity, as recommended by the FDA.15 Secondary vasomotor endpoints included mean change from baseline to all nonprimary time points in moderate-to-severe hot flush rate and hot flush severity, time to first 100% reduction from baseline in daily moderate-to-severe hot flush rate, and percent of participants who achieved a 50% to 100% reduction from baseline to all postbaseline time points in daily moderate-to-severe hot flush rate and severity. Analyses were conducted on the intent-to-treat population, defined as all randomly assigned participants who received study drugs and provided diary responses for at least 1 full day during the treatment period, using a last observation carried forward (LOCF) approach. For vulvovaginal atrophy symptoms, coprimary efficacy endpoints were mean change from baseline to week 12 (or last visit for participants who discontinued prematurely) in 1) the moderate-to-severe vulvovaginal symptom identified by the participant as being the most bothersome, 2) vaginal pH, and 3) vaginal maturation index. Participants were included in the primary analysis of the respective endpoint if at screening they 1) self-identified at least one moderate-to-severe vulvovaginal symptom that was most bothersome to them, 2) had vaginal pH more than 5.0, and 3) had 5% or fewer superficial cells in the vaginal wall specimen. Secondary endpoints included change from baseline to all postbaseline time points during double-blind treatment in participant vaginal health self-assessment and change from baseline to week 12 in physician assessment of vaginal atrophy. Other efficacy assessments included participant global efficacy questions, Utian Quality of Life16 and Menopause Quality of Life17 questionnaires, and a participant opinion survey. Analyses were conducted on the intent-to-treat population using observed data.
Safety evaluations at baseline and end of treatment comprised clinical laboratory tests, including hematology, chemistry, lipid profile, urinalysis, coagulation profile, and C-reactive protein; vital signs; and physical and gynecologic examination, including endometrial biopsy (or endometrial thickness assessment by transvaginal ultrasonography if insufficient tissue was obtained). Blood samples for trough serum concentrations of E2, estrone, estrone sulfate, and sex hormone binding globulin (Esoterix, Inc., Calabasas, CA) were collected at baseline and each study visit.
The planned sample size (127 participants per treatment group for the placebo and E2 0.87 g/d and 1.7 g/d treatment groups) was calculated (using a two-tailed calculation) to detect a difference of 2.0 or more between treatment groups in mean change from baseline at week 4 and week 12 in daily moderate-to-severe hot flush frequency, based on an estimated standard deviation of change from baseline of 5.0 hot flushes per day and an 80% power to detect differences at the .05 level. For the 2.6 g/d dose, a sample size of 50 participants was determined to detect a difference of 3.0 from the placebo group at week 12. The study was not powered for vaginal atrophy assessments. For all efficacy variables an adjustment for multiple testing was performed using Dunnett’s test. Before the implementation of parametric methods of analysis, the distribution of primary analysis variables was examined to determine whether the model assumptions were satisfied (normal distribution by the Wilk-Shapiro test and constant variance), and transformations were used if warranted. Model assumptions were fulfilled in all cases and nonparametric statistics were not required in the analysis. An analysis of covariance model was used for the primary efficacy measures with covariates including center (sites enrolling small numbers were pooled), treatment, covariate-by-treatment interaction, and center-by-treatment interaction, where the covariate was the baseline value of the variable being analyzed. Comparison of the treatment groups with respect to other quantitative variables were based on a two-way analysis of variance model, including factors for center, treatment, and center-by-treatment interaction. For either model, the interaction term was removed from the final model if not statistically significant (P>.10). Each dose group was compared with placebo using Dunnett’s test, with comparisons based on the least squares means derived from the analysis of covariance or analysis of variance. Comparability of treatment groups with respect to categorical variables was based on the Cochran-Mantel-Haenszel general association statistic with center as the stratification factor followed by the Cochran-Mantel-Haenszel test if a statistically significant treatment group difference was found. Comparability of treatment groups with respect to time to 100% reduction in daily moderate-to severe hot flushes was based on a log rank test and Kaplan-Meier survival curve estimates. Unless otherwise indicated, statistical significance was declared if the 2-tailed P value was .05 or less. All statistical analyses were performed using SAS 8.2 (SAS Institute Inc., Cary, NC).
This study was conducted over a 30-month period (September 2003 to April 2005) and involved screening of 1,971 participants, 484 of whom were subsequently randomly assigned to double-blind treatment (Fig. 1). Most participants in each treatment group completed the study, including 97% in the low-dose (0.87 g/d) E2 gel group and 93% to 94% in the other groups, respectively. Randomly assigned participants were between 28 and 74 years of age and had been menopausal for approximately 5 months to 43 years. Demographic and baseline characteristics for participants were not significantly different (P>.05) across treatment groups (Table 1). Most participants were white, and most had received prior hormone therapy, including prescription estrogen therapy, for treatment of hot flushes. At baseline, mean serum E2 levels ranged between 9.7 and 13.2 pg/mL across treatment groups, and mean hot flush rate ranged between 12.9 and 13.5 hot flushes per day, with a mean severity of 2.4 in all treatment groups (Table 1).
With increasing E2 doses, mean trough serum E2 increased from 17 to 29 pg/mL over the 12-week treatment period in participants receiving E2 gel (Fig. 2). Consistent with the increased serum E2, the rate and severity of hot flushes decreased. By weeks 3–5, E2 gel reduced the moderate-to-severe hot flush rate by at least seven hot flushes per day (P<.001) (Fig. 3). Statistically significant differences from placebo in mean change from baseline hot flush rate started at week 3 for participants receiving either 1.7 g/d (–8.2 compared with –5.4; P=.007) or 2.6 g/d E2 gel (–9.5 compared with –5.4; P<.001) and at week 5 for participants receiving 0.87 g/d E2 gel (–7.7 compared with –5.5; P<.001), approaching significance at the week 4 coprimary endpoint (–6.6 compared with –5.4; P=.071) (Fig. 3, upper panel). Statistically significant differences were maintained in all E2 dose groups through the week 12 coprimary endpoint even though the placebo response persisted throughout the study also (P<.001) (Fig. 3, upper panel). Similar results were seen for hot flush severity (Fig. 3B). All secondary efficacy measures of vasomotor symptoms favored transdermal E2 treatment. Significantly higher percentages of participants receiving E2 gel than placebo had at least a 50% reduction from baseline in daily moderate-to-severe hot flush rate at week 4 (P<.001) (Fig. 4A). At week 12, when the majority of participants in each E2 gel treatment group had achieved at least a 50% reduction, fewer than half of those in the placebo group had done so (P<.001); significantly higher percentages of E2 gel participants had also achieved at least an 80%, 90%, 95%, and 100% reduction (Fig. 4, lower panel). Based on these data, the numbers needed to treat for benefit (1/[percent E2 participants–percent placebo participants] with 95% confidence intervals) for the 0.87 g/d group were 3.2 (2.4–5.0), 4.2 (2.9–7.3), 4.5 (3.2–8.0), and 6.3 (4.0–13.8), respectively for a decrease of 80%, 90%, 95%, and 100% in hot flush number, whereas these same numbers for the 2.6 g/d group were 1.7 (1.4–2.2), 1.8 (1.5–2.3), 1.8 (1.5–2.3) and 2.3 (1.8–3.3) at week 12.
The effects of this E2 gel on vulvovaginal atrophy symptoms are presented in Table 2. Statistically significant (P<.001) improvements in two of the three primary measures of vulvovaginal atrophy were observed for all doses of E2 gel relative to placebo, with the vaginal pH being more acidic and the vaginal maturation index indicating a more mature cellular pattern (Table 2). The lowest dose of E2 gel also significantly improved participants’ most bothersome vulvovaginal atrophy symptom (P<.05) (Table 2). Physician assessments of vulvovaginal atrophy, which indicated generally mild (ie, score of approximately 1) or very mild severity (score less than 1) across treatment groups at baseline, showed significant mean reductions from baseline to week 12 for each dose of E2 gel relative to placebo in vaginal atrophy (0.56, 0.72, and 0.65 compared with 0.19; P<.001), pallor (0.56, 0.64, and 0.68 compared with 0.15; P<.001), and dryness (0.79, 0.88, and 0.88 compared with 0.27; P<.001) and for the 0.87 g/d dose of E2 gel only, in vaginal tissue petechiae (0.31 compared with 0.09; P<.01). Likewise, participant self-assessment of vulvovaginal atrophy symptoms also indicated generally mild or very mild severity across treatment groups at baseline, but significant mean reductions from baseline to various time points were noted for the lowest dose of E2 gel relative to placebo in vaginal dryness (1.03 compared with 0.74 at week 12; P=.030) and pain or difficulty passing urine (0.92 compared with 0.56 at week 12; P=.016).
Other efficacy assessments also favored this E2 gel. For the 0.87 g/d, 1.7 g/d, and 2.6 g/d E2 gel doses compared with placebo, a greater proportion of participants felt that E2 gel produced great improvement in their hot flushes (61%, 76%, and 77% compared with 27%) and either moderate or great improvement in vulvovaginal atrophy (53%, 57%, and 69% compared with 21%). Relative to placebo, E2 gel significantly improved in a dose-related manner the vasomotor (3.2, 3.9, and 4.0 compared with 1.5; P<.001), psychosocial (0.7, 0.8, and 0.9 compared with 0.3; P<.05 or <.01), and physical (0.7, 0.9, and 1.1 compared with 0.4; P<.05 or <.001) domain scores of the Menopause Quality of Life. The higher doses of E2 gel (1.7 g/d and 2.6 g/d) also improved the sexual factor score of the Utian Quality of Life (0.5 and 1.2 compared with –0.3; P<.05 and <.001), and the 1.7 g/d dose improved the emotional factor (0.9 compared with 0.2; P<.05) and overall score (2.3 compared with 1.1; P<.05).
The overall incidence of treatment-emergent adverse events, defined as the percentage of patients in each group with any adverse event on study drug, increased with increasing dose of E2 gel to 59%, 65% and 68% for the 0.87 g/d, 1.7 g/d, and 2.6 g/d treatment groups, respectively, and was slightly higher than the incidence in the placebo group (56%). Reproductive system and breast disorders were the most common treatment-emergent adverse events and significant between-treatment differences in incidences were observed (P<.001). They included breast tenderness (6.6%, 7.7%, and 21.7% compared with 3.6%), metrorrhagia (4.4%, 9.2%, and 20.7% compared with 2.2%), vaginal discharge (1.5%, 3.5%, and 7.2% compared with 0.7%), endometrial hypertrophy (0%, 0.7%, and 7.2% compared with 0%), and nipple pain (0.7%, 2.1%, and 5.8% compared with 0%). Among other adverse events, only nausea (in two placebo and 10 E2 gel participants) and peripheral edema (in three E2 gel participants) were considered treatment related. The E2 gel was well tolerated at the site of application, with dryness (1% to 4% across treatment groups) and erythema (2% to 3%) being the most common application site disorders.
Eight participants (one placebo, one E2 0.87 g/d, four E2 1.7 g/d, and two E2 2.6 g/d) prematurely discontinued due to treatment-related breast swelling, endometrial hypertrophy, metrorrhagia, migraine, limb discomfort, ovarian mass, rash, peripheral edema, or increased weight. At endpoint, 5 participants (all E2 gel 2.6 g/d) had simple hyperplasia without atypia, and one participant (1.7 g/d) had simple hyperplasia with atypia, which, after a curettage, final pathology revealed as simple hyperplasia without atypia. Three participants had clinically significant laboratory values, including one participant (E2 2.6 g/d) with low hematocrit (31.4%), hemoglobin (10.4 g/dL), and red blood cell (3.29×106/μL) values, and two participants (1 E2 0.87 g/d and 1 E2 2.6 g/d) with low protein S antigen free activity (55% and 56%), and these laboratory findings were not considered to be clinically important by the Investigator. Estradiol gel had no clinically significant effects on vital signs, body weight, or physical examinations. Mean sex hormone binding globulin concentrations were increased slightly over placebo only at the highest E2 gel dose (111 nM compared with 87 nM at week 12; P<.01).
In this study, the two higher doses of a new transdermal E2 gel decreased vasomotor symptoms significantly before the predefined efficacy landmark time point (4 weeks) consistent with current FDA guidance for establishing efficacy for this indication.15 The 0.87 g/d dose showed its significant effect at week 5. By virtue of not being effective at the predesignated, landmark 4 week time point, no dose lower than 0.87 g/d would likely be effective at 4 weeks in a similar study population. This dose, which provides a nominal daily delivery of 0.0125 mg E2, provided significant mean reductions in hot flush rate and severity continued through the end of study. Beneficial effects on vasomotor symptoms were accompanied by statistically significant improvements in objective (vaginal pH, vaginal maturation index) and subjective (most bothersome vulvovaginal symptom) measures of vaginal atrophy symptoms at endpoint. Participants receiving the lowest dose of this E2 gel were also significantly more likely than placebo participants to have at least a 50% reduction in hot flush severity at week 4, and at least a 50% through 100% reduction at week 12. They also had greater global improvements in hot flush and vulvovaginal atrophy symptoms, a better quality of life as assessed by the vasomotor factor, psychosocial factor, and physical factor domains of the Menopause Quality of Life, and less vaginal dryness and pain or difficulty passing urine. Physician assessments of vulvovaginal atrophy agreed with participant assessments, with vaginal atrophy, vaginal pallor, vaginal dryness, and vaginal tissue petechiae significantly improved by the lowest dose of E2 gel.
In large part, the beneficial effects of the lowest dose of this E2 gel also were seen at the 1.7 g/d and 2.6 g/d doses, indicating that participants who fail to respond to treatment at the lowest dose may derive relief of their symptoms from higher doses. Nevertheless, the 0.87 g/d dose provided the lowest exposure to E2 that was of value to a representative population of symptomatic participants over time and resulted in fewer of the well-known, dose-dependent adverse effects of estrogen treatment.
In individual patients a lower daily dose of E2 might be effective in treating these symptoms, especially if they are willing to wait for a longer period of time, likely greater than 5 weeks. However, for the lowest Elestrin (BioSante Pharmaceuticals, Inc.) dose studied, 0.87 g/d, Figure 4 shows that when compared with placebo only a small fraction of participants reported a 100% reduction in hot flush number by 4 weeks, suggesting that the probability of a lower E2 dose relieving hot flushes to this extent by 4 weeks is low. In this context, to date there is no FDA-approved E2 dose lower than 0.025 mg E2 available for treatment of hot flushes.
A number of authoritative groups specify that the lowest dose of estrogen should be used for the shortest period of time, including the American College of Obstetrics and Gynecology, North American Menopause Society, and FDA.18–20 There are potential benefits that may occur with a low dose of E2 for the treatment of menopausal symptoms. First, treatment compliance may increase. The incidence of breast tenderness and of vaginal bleeding was low in participants receiving the 0.87 g/d E2 gel dose. Because these are both considered to be unacceptable side effects leading to discontinuation of treatment in patients undergoing therapy,21 use of a low dose may increase the chance that patients will continue treatment for their menopausal symptoms.10,22
The current study did not investigate estrogen treatment for longer than 12 weeks, nor did it investigate timing or doses of progestin therapy for endometrial protection. However, a recent publication reported that a transdermal E2 patch preparation that delivers a nominal daily dose of 0.014 mg E2 did not cause endometrial hyperplasia after a year of therapy and only a slight increase in vaginal bleeding after 2 years.23 The product label for this compound now approved for treatment of osteoporosis but not hot flushes recommends 14 days of progestin every 6–12 months, with yearly endometrial biopsies.23,24 None of the participants in the study population described in this article who received the lowest dose of E2 gel, with a nominal daily dose of 0.0125 mg E2, had endometrial hyperplasia or any other suspicious endometrial findings. In contrast, the highest dose did show hyperplasia, consistent with the dose-dependent effect of E2 on the endometrium as seen in other 12-week studies.25 If low-dose formulations, such as the 0.87 g/d dose of this gel, are shown not to cause endometrial hyperplasia after extended treatment, less frequent than monthly progestin use could be possible and might even prove beneficial. However, at this point, monthly progestin therapy is the current class labeling to prevent endometrial hyperplasia and endometrial adenocarcinoma, findings that have been observed in other 12-week studies of unopposed estrogen therapy.
The WHI showed that the relative risk of breast cancer was elevated in patients receiving a combination of estrogen and progestin but not in patients receiving estrogen only. If the progestin component increases breast cancer risk, as has been suggested,26,27 and if there are reproducible data showing the endometrial safety of these low daily E2 doses, it could be speculated that the need for less progestin after a lowest effective E2 dose may lessen the risk of breast cancer in estrogen-treated women. This, however, has not been studied.
The strength of the current investigation is that it is a randomized, double-blind, placebo-controlled, multicenter study, although a potential weakness may be the block design in site allocation of study drug that has a theoretic risk for partial unblinding.
Taken together, the results of this study support the 0.87 g/d dose of Elestrin (BioSante Pharmaceuticals, Inc.) as a lowest effective estrogen dose in this population and may prove to be of value to postmenopausal women who require treatment for their vasomotor and vulvovaginal symptoms.
1. Sikon A, Thacker HL. Treatment options for menopausal hot flashes. Cleve Clin J Med 2004;71:578–82.
2. Treatment of menopausal vasomotor symptoms. Med Lett Drugs Ther 2004;46:97–102.
5. Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C, Stefanick ML, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women’s Health Initiative randomized controlled trial. JAMA 2002;288:321–33.
6. Shumaker SA, Legault C, Rapp SR, Thal L, Wallace RB, Ockene JK, et al. Estrogen plus progestin and the incidence of dementia and mild cognitive impairment in postmenopausal women: the Women’s Health Initiative Memory Study: a randomized controlled trial. JAMA 2003;289:2651–62.
7. Shumaker SA, Legault C, Kuller L, Rapp SR, Thal L, Lane DS, et al. Conjugated equine estrogens and incidence of probable dementia and mild cognitive impairment in postmenopausal women: Women’s Health Initiative Memory Study. JAMA 2004;291:2947–58.
8. Anderson GL, Limacher M, Assaf AR, Bassford T, Beresford SA, Black H, et al. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women’s Health Initiative randomized controlled trial. JAMA 2004;291:1701–12.
9. Hsia J, Langer RD, Manson JE, Kuller L, Johnson KC, Hendrix SL, et al. Conjugated equine estrogens and coronary heart disease: The Women’s Health Initiative. Arch Intern Med 2006;166:357–65.
10. Lobo RA. The rationale for low-dose hormonal therapy. Endocrine 2004;24:217–21.
11. Ettinger B. Vasomotor symptom relief versus unwanted effects: role of estrogen dosage. Am J Med 2005;118:74S–8S.
12. Grady D, Sawaya GF. Discontinuation of postmenopausal hormone therapy. Am J Med 2005;118 suppl:163–5.
13. Center for Drug Evaluation and Research. Guidance for industry: noncontraceptive estrogen drug products for the treatment of vasomotor symptoms and vulvar and vaginal atrophy symptoms—recommended prescribing information for health care providers and patient labeling. Nov Labeling Revision 4 (draft). Rockville (MD): US Department of Health and Human Services, Food and Drug Administration; 2005.
14. McKenzie J, Jaap AJ, Gallacher S, Kelly A, Crawford L, Greer IA, et al. Metabolic, inflammatory and haemostatic effects of a low-dose continuous combined HRT in women with type 2 diabetes: potentially safer with respect to vascular risk? Clin Endocrinol (Oxf) 2003;59:682–9.
15. Center for Drug Evaluation and Research. Guidance for industry: estrogen and estrogen/progestin drug products to treat vasomotor symptoms and vulvar and vaginal atrophy symptoms—recommendations for clinical evaluation. Jan (draft). Rockville (MD): US Department of Health and Human Services, Food and Drug Administration; 2003.
16. Utian WH, Janata JW, Kingsberg SA, Schluchter M, Hamilton JC. The Utian Quality of Life (UQOL) Scale: development and validation of an instrument to quantify quality of life through and beyond menopause. Menopause 2002;9:402–10.
17. Hilditch JR, Lewis J, Peter A, van Maris B, Ross A, Franssen E, et al. A menopause-specific quality of life questionnaire: development and psychometric properties. Maturitas 1996;24:161–75.
18. Vasomotor symptoms. Obstet Gynecol 2004;104:106S– 117S.
19. North American Menopause Society. Treatment of menopause-associated vasomotor symptoms: position statement of the North American Menopause Society. Menopause 2004;11:11–33.
21. den Tonkelaar I, Oddens BJ. Determinants of long-term hormone replacement therapy and reasons for early discontinuation. Obstet Gynecol 2000;95:507–12.
22. Continuation rates for oral contraceptives and hormone replacement therapy. The ESHRE Capri Workshop Group. Hum Reprod 2000;15:1865–71.
23. Johnson SR, Ettinger B, Macer JL, Ensrud KE, Quan J, Grady D. Uterine and vaginal effects of unopposed ultralow-dose transdermal estradiol. Obstet Gynecol 2005;105:779–87.
24. Prestwood KM, Kenny AM, Kleppinger A, Kulldorff. Ultralow-dose micronized 17beta-estradiol and bone density and bone metabolism in older women: a randomized controlled trial. JAMA 2003;290:1042–8.
25. Nelson HD. Commonly used types of postmenopausal estrogen for treatment of hot flashes: scientific review. JAMA 2004;291:1610–20.
26. Hulley SB, Grady D. The WHI estrogen-alone trial—do things look any better? JAMA 2004;291:1769–71.
27. Beral V, Million Women Study Collaborators. Breast cancer and hormone-replacement therapy in the Million Women Study [published erratum appears in Lancet. 2003;362:1160]. Lancet 2003;362:419–27.