Vulvar and vaginal atrophy (VVA), a chronic postmenopausal health condition that occurs because of a hypoestrogenic state,1,2 is common yet often underreported.1-3 Approximately one third of a woman’s lifetime is spent in the postmenopausal period,4,5 with a high probability of experiencing one or more hormone deficiency–related problems such as VVA. Unlike some symptoms of menopause, such as hot flushes, that lessen or disappear with time, dyspareunia and urogenital symptoms of VVA (eg, vaginal dryness, itching, and irritation) usually persist and can even worsen without treatment.6 In addition to physical discomfort and sexual dysfunction, VVA may also be associated with significant emotional distress and reduced quality of life.6-8 Moreover, the condition has the potential to negatively affect the lives of a woman and her sexual partner.9 Among postmenopausal women who have never used menopausal hormone therapy, the incidence of VVA symptoms, including dyspareunia, reaches up to approximately 60%.7,10 Postmenopausal women affected by VVA equally include those who are sexually active and those who are not.7 Despite its potentially negative impact on sexual function, psychosocial well-being, and partner relationships, VVA frequently goes untreated.10
Currently available prescription treatment options for VVA are orally or vaginally administered estrogen-based agents; their use is limited, and they are administered at low doses because of the potential for associated systemic estrogenic effects, especially with long-term use.10,11 In the wake of the publication of the Women’s Health Initiative estrogen-progestin study in 2002,12 safety concerns led many women to discontinue hormone therapy. In a subsequent study of women who stopped taking hormone therapy after the Women’s Health Initiative, 30% reported the return of bothersome symptoms a median of 1 week after stopping hormone therapy.13 The stimulatory effects of estrogens on breast and endometrial tissues also have directed the use of estrogen preparations at the lowest effective dose for the shortest duration of time, consistent with treatment goals,10,14 and have prompted interest in treatment alternatives such as selective estrogen receptor modulators (SERMs).15 SERM treatments have been developed to prevent and treat diverse conditions such as breast cancer, postmenopausal osteoporosis, and postmenopausal VVA.15
Tissues potentially targeted by SERMs include genitourinary, uterine, breast, and bone tissues.9,15,16 Depending on the tissue’s expression pattern, conformations of estrogen receptor subtypes, and effects of regulator proteins, each SERM has a unique profile in its effects across these tissues.9,15-17 Tissue selectivity is a key characteristic of SERMs that affects both their efficacy and their safety, and ultimately their viability as a treatment. An ideal SERM for treating VVA should have agonist effects on vaginal tissue, but neutral effects on the breast and endometrium.18 Several investigational SERMs have been studied; however, to date, none has been approved for the treatment of VVA and its associated symptoms or physiological changes.
Ospemifene, a novel SERM, is an oral compound undergoing investigation for the treatment of VVA due to menopause, including moderate to severe dyspareunia and physiological changes (parabasal cell, superficial cell, and pH). Ospemifene has been reported in another clinical trial to be effective and well tolerated in postmenopausal women with VVA.10 The safety of ospemifene has been studied for up to 1 year in two long-term studies.19,20 Based on a search of the literature to date, this work reports the largest dyspareunia cohort (n = 605) ever studied for the treatment of postmenopausal women with VVA.
Two identically designed clinical studies were conducted in parallel for each of the two major symptoms of VVA: dyspareunia and vaginal dryness. Both studies assessed the efficacy, safety, and tolerability of ospemifene 60 mg/day—one in the treatment of moderate to severe dyspareunia, and the other in the treatment of moderate to severe vaginal dryness. For each of the two symptoms, separate randomization schedules, participant databases (locked separately), and data analyses were conducted. In this publication, only the data on the efficacy and safety of ospemifene for the treatment of dyspareunia are discussed; data for the vaginal dryness group (n = 314) will be reported separately.
Study design and population
This phase 3 study (NCT00729469) used a randomized, double-blind, parallel-group design to compare the efficacy, safety, and tolerability of ospemifene 60 mg/day versus placebo for treating moderate to severe dyspareunia in postmenopausal women with VVA (Fig. 1). A visual examination of the vagina was performed to assess petechiae, pallor, friability, vaginal dryness, and redness on a 4-point scale (0, none; 1, mild; 2, moderate; 3, severe).
Potential participants completed a questionnaire in which they were asked to self-report their most bothersome symptom (MBS) on a 4-point scale (0, none; 1, mild; 2, moderate; 3, severe). Those who reported having moderate or severe vaginal pain (dyspareunia) with sexual activity as their MBS continued to the screening phase of the study. Participants reporting an MBS other than dyspareunia were not included in this study.
For the treatment phase, a total of 605 women were randomized to ospemifene 60 mg/day (n = 303) or placebo (n = 302) and took a once-daily dose of study medication with food in the morning for 12 weeks. During the treatment period, participants were seen on weeks 4 and 12 for completion of the VVA symptom questionnaire, assessment of vaginal pH, vaginal smear, and visual examination of the vagina. Also on week 12, participants underwent transvaginal ultrasound (TVU) and endometrial biopsy.
For blinding, both ospemifene 60 mg and placebo were supplied as tablets that were identical in appearance. In addition, based on a discussion with the US Food and Drug Administration, a nonhormonal vaginal lubricant was provided to all participants. Participants were instructed to use the lubricant as needed to provide them on-demand comfort during sexual activity and to help understand the effects of pharmacological treatments with background use of over-the-counter lubricants—a common practice in this treatment population. Lubricant use was recorded by study participants on a daily diary.
The investigation was conducted at 110 sites in the United States, beginning with the first participant completed on August 4, 2008, and ending with the last participant completed on July 30, 2009. The protocol was reviewed and approved by the Copernicus Group Institutional Review Board (Durham, NC) for most clinical sites; local institutional review boards were used for the remaining sites. The study was conducted in accordance with the Declaration of Helsinki and current Good Clinical Practice.
Study participants were aged 40 to 80 years at randomization and were postmenopausal (defined as at least 12 mo since the last spontaneous menstrual bleeding, at least 6 weeks since bilateral oophorectomy, or hysterectomized women with intact ovaries and serum follicle-stimulating hormone levels >40 IU/L). Study participants were diagnosed as having VVA, defined as 5% or less superficial cells in the maturation index (MI) of the vaginal smear and a vaginal pH higher than 5. Participants enrolled in this study either were hysterectomized or had an intact uterus with a double-layer endometrial thickness less than 4 mm (determined by centrally read ultrasound) and had no evidence of hyperplasia, cancer, or other pathology (determined by endometrial biopsy). To be included, women must have had a negative Papanicolaou test result or lacked an intact cervix. Study participants also must have had a negative mammogram result 9 months or less before randomization and a normal breast examination result at screening. All randomized study participants provided a written informed consent form.
Women were excluded if they had a body mass index of 37 kg/m2 or higher, a systolic blood pressure of 180 mm Hg or higher, or a diastolic blood pressure of 100 mm Hg or higher, or if they presented with any clinically significant abnormal gynecological findings, other than signs of vaginal atrophy, such as the following: uterine bleeding of unknown origin, uterine polyps or symptomatic and/or large uterine fibroids (>3 cm), or vaginal infection requiring medication. Any clinically significant abnormal findings from the physical examination, mammography, electrocardiogram, safety laboratory tests, or liver function screening excluded women from participation in the study. Women were also excluded if they consumed more than 14 alcoholic drinks per week; took heparin, digitalis alkaloids, or strong cytochrome P450 3A4 inhibitors; used any hormonal medications, SERMs, or products expected to have estrogenic and/or antiestrogenic effects within prespecified time frames before study screening (eg, 14 d for vaginal hormonal products and 60 d for oral or transdermal estrogen and/or progestin therapy); or used ospemifene before study screening. Exclusion criteria also included women who were positive for factor V Leiden mutation or had current or past cerebrovascular incidents, thromboembolic disorders, blood coagulation disorders, severe hepatic or renal impairment, or suspicion of malignancy on mammography within 10 years.
Efficacy analysis included the change from baseline to week 12 (primary efficacy endpoints) and week 4 (secondary efficacy endpoints) in the following four coprimary endpoints: percentage of parabasal cells and percentage of superficial cells in the MI of the vaginal smear, vaginal pH, and severity of the MBS (ie, dyspareunia associated with sexual intercourse). Parabasal cells, superficial cells, and pH were evaluated using an analysis of covariance model. Within this model, the response variables were the change from baseline to week 4 or week 12; for missing values, the last observation was used (the last observation carried forward [LOCF]). The baseline value was the covariate, and treatment and study center were the fixed effects. MBS severity was scaled as none (0), mild (1), moderate (2), or severe (3), and the change from baseline in MBS severity was analyzed using the Cochran-Mantel-Haenszel row mean scores test, controlling for the study center.
Secondary endpoints—including change in the severity of dyspareunia for week 4 observed cases and responders on week 12/LOCF—were analyzed inferentially using a Cochran-Mantel-Haenszel row mean scores test, controlling for the study center. For changes from baseline in MI and Female Sexual Function Index total and domain scores, an analysis of covariance model was applied as described previously.
Two sets of participant populations were analyzed. The primary analysis set was the intent-to-treat (ITT) population, which consisted of all randomized participants who took at least one dose of the study medication. A supportive analysis set was the per-protocol (PP) population, which consisted of all participants who had completed at least 10 weeks of treatment and the end-of-study assessments, had taken 85% or more of the study medication, had no other major protocol violations, and did not have a vaginal infection or any medical condition that would confound the primary efficacy assessment. To show the effectiveness of ospemifene for the treatment of VVA, we required all coprimary variables (percentage of parabasal cells, percentage of superficial cells, vaginal pH, and dyspareunia) in the ITT analysis to demonstrate statistically significant improvement compared with placebo.
Safety analyses were conducted on the ITT population only. At the time of informed consent, study participants were asked to report spontaneously all adverse events (AEs) throughout the study period. Participants were queried about AEs at every visit after screening. AEs were coded using the Medical Dictionary for Regulatory Activities (MedDRA, version 10.1), and treatment-emergent AEs (TEAEs) were tabulated by treatment group, system organ class, preferred term, causality, and severity. Safety was also assessed by physical examination, breast palpation, cervical Papanicolaou test, clinical laboratory analyses, centrally read 12-lead electrocardiogram, vital signs, weight, and treatment compliance.
At screening and on week 12 for participants with an intact uterus, endometrial safety was assessed by TVU of endometrial thickness and histology via biopsy to determine the incidence of serious endometrial outcomes. Double-layer endometrial thickness was measured locally by a trained study staff member and by a central reader for the screening and week 12 assessments; in addition, local TVU measurements were conducted during and after the treatment period to confirm the presence of uterine polyps or other potentially important clinical findings. The histological characteristics of the endometrium were ascertained by an evaluation of endometrial biopsy specimens obtained at screening and 12 weeks (or at the end of treatment). Biopsies were assessed according to predefined and generally accepted microscopic criteria (Blaustein’s classification). Analysis of biopsy samples was conducted at a central laboratory. Slides were reviewed by two independent pathologists who were blinded to the study treatment and to each other’s assessments. In cases of disagreement between the two, a third pathologist reviewed the sample.
Participant disposition and demographics
A total of 605 postmenopausal women were included in the ITT population, with 303 randomized to the ospemifene 60 mg group and 302 randomized to the placebo group. The PP population included a total of 506 participants (255 and 251 in the ospemifene and placebo groups, respectively). Overall, 544 (89.9%) completed the 12-week study, and 61 (10.1%) discontinued. Of those who discontinued, 8.3% were in the ospemifene group and 11.9% were in the placebo group (Fig. 2). The most common reason for discontinuation were AEs, occurring in 4.6% of participants in the ospemifene group compared with 3.0% in the placebo group. The baseline characteristics and baseline values for the coprimary endpoints were similar between the study groups (Tables 1 and 2). Most participants were white (90.6%), were aged from 40 to 79 years, and had body mass index values ranging from 16.7 to 37.1 kg/m2.
Primary efficacy analysis
In the ITT analysis (n = 605), ospemifene 60 mg/day demonstrated statistically significant efficacy compared with placebo in the mean change from baseline for each of the four coprimary endpoints. After 12 weeks of treatment, the percentage of parabasal cells significantly decreased by 40.2% in the ospemifene group compared with no reduction in the placebo group (P < 0.0001; Fig. 3). Similarly, the percentage of superficial cells significantly increased by 12.3% in the ospemifene group compared with 1.7% in the placebo group (P < 0.0001; Fig. 4). During the same period, the mean reduction in vaginal pH in the ospemifene group (−0.94) was significantly greater than that in the placebo group (−0.07; P < 0.0001; Fig. 5). The reduction in MBS severity score was also significantly different between the ospemifene group (−1.5) and the placebo group (−1.2; P = 0.0001; Fig. 6), demonstrating efficacy in the treatment of VVA among the participants reporting dyspareunia as their MBS.
Notably, the percentage of participants reporting no vaginal pain or mild vaginal pain with sexual activity on week 12 was greater in the ospemifene group (38.0% or 25.1%, respectively) than in the placebo group (28.1% or 19.2%, respectively). Also, the severity of vaginal pain on week 12 in the ITT population improved by two to three levels in 52.8% of participants in the ospemifene group compared with 38.8% of participants in the placebo group (a three-level improvement being a change from “severe” to “none”; a two-level improvement being either a change from “severe” to “mild” or from “moderate to “none”).
In the PP population analysis (n = 506), ospemifene 60 mg/day demonstrated statistically significant efficacy compared with placebo in the mean change from baseline for each of the four coprimary endpoints. After 12 weeks of treatment, the percentage of parabasal cells and superficial cells significantly changed in the ospemifene group (decreased by 42.1% and increased by 13.2%, respectively) compared with the placebo group (decreased by 0.2% and increased by 1.9%, respectively; P < 0.0001; Figs. 3 and 4). Also, the reduction in vaginal pH in the ospemifene group (−1.1) was significantly greater than that observed in the placebo group (−0.08; P < 0.0001; Fig. 5), as was the reduction in MBS severity score (ospemifene −1.6 vs placebo −1.2; P = 0.0004; Fig. 6). Furthermore, within the PP population, the percentage of participants reporting no vaginal pain and mild vaginal pain with sexual activity on week 12 was greater in the ospemifene group (38.8% and 27.1%, respectively) than in the placebo group (29.5% and 21.5%, respectively). On week 12, 55.7% of participants in the ospemifene group changed their rating of vaginal pain severity from “severe” to “none” (a three-level improvement) or from “severe” to “mild” or from “moderate” to “none” (a two-level change) compared with 41.8% of participants in the placebo group.
Secondary efficacy analysis
A secondary analysis of the coprimary endpoints in the ITT population on week 4 demonstrated differences between ospemifene 60 mg/day and placebo in the mean change from baseline in MI and vaginal pH. Ospemifene 60 mg/day demonstrated efficacy compared with placebo in the percentage reduction in parabasal cells (−38.0% vs 0.3%, respectively) and in the percentage increase in superficial cells (13.0% vs 1.8%, respectively). Similarly, the reduction in vaginal pH on week 4 was different between the ospemifene group and the placebo group (−0.82 vs −0.15). A trend toward better improvement in vaginal pain associated with sexual activity in the ospemifene group compared with the placebo group was noted at 4 weeks as well, although the difference between the study groups was not statistically significant (P = 0.1698).
Each of the five parameters assessed by visual examination of the vagina showed the ospemifene group at 12 weeks to have a greater number of participants with an improvement of two to three levels from baseline (a three-level improvement being a change from “severe” to “none”; a two-level improvement being a change from “severe” to “mild” or from “moderate” to “none”) compared with the placebo group (petechiae, 10.4% vs 6.3%; pallor, 21.3% vs 10.0%; friability, 13.2% vs 7.7%; vaginal redness, 12.1% vs 8.2%; vaginal mucosal dryness, 36.6% vs 13.7%).
Use of lubricant
During the first week of treatment, similar proportions of participants used lubricants (41.7% and 43.1% in the ospemifene and placebo groups, respectively). On week 12, the percentage of women using lubricants somewhat decreased for both groups, but more so in the ospemifene group (35.1%) compared with the placebo group (39.3%). In contrast, the frequency of sexual activity remained consistent across the study groups from week 1 to week 12.
Safety and tolerability
More participants in the ospemifene group (91.7%) than in the placebo group (88.1%; Fig. 2) completed the study. The percentage of participants who discontinued the study because of an AE was higher for the ospemifene group (4.6%) than for the placebo group (3.3%); 3.3% and 1.3% of participants discontinued because of a treatment-related AE, respectively (Table 3).
Of 605 participants, 186 (61.4%) in the ospemifene group and 154 (51.0%) in the placebo group experienced at least one TEAE; most were considered mild to moderate. Most TEAEs were classified as not related to the study drug or unlikely to be related to the study drug. Those considered related to the study drug were reported in 79 (26.1%) participants in the ospemifene group and in 44 (14.6%) participants in the placebo group. The most frequently reported treatment-related AEs were hot flushes, reported by 6.6% of participants in the ospemifene group and 3.6% of participants in the placebo group. However, hot flushes led to treatment discontinuation for only one participant in the ospemifene group and for one participant in the placebo group. The other TEAEs that occurred in at least 3% of participants in the ospemifene group were urinary tract infection, vaginal candidiasis, vaginal discharge, vulvar and vaginal mycotic infections, nasopharyngitis, and headache (Table 4).
The incidence of serious AEs was the same for the ospemifene and placebo groups (four participants [1.3%] for both), with no serious AEs considered related to the study drug. Vaginal bleeding was not reported in any of the participants in the ospemifene group and in two (0.7%) participants in the placebo group. In the placebo group, one (0.3%) participant experienced a cerebrovascular event (ischemic stroke).
Endometrial thickness was assessed by centrally read TVU at screening and 12 weeks. Within each treatment group, one participant was suspected to have a polyp based on centrally read TVU set at a low threshold; therefore, local “live” reads were used for confirmation. The presence of a polyp was confirmed by a local read for the placebo participant but was found to be benign upon excision. A local read of the treatment participant, however, returned a normal result; thus, the polyp was not confirmed. Ospemifene caused a slight increase in endometrial thickness (Table 5). Endometrial histology assessments showed two (1.4%) cases of active proliferation in the ospemifene group, which resolved after the study; no cases were observed in the placebo group. At 12 weeks, no cases of endometrial hyperplasia, polyps, or cancer were observed in the endometrial biopsy samples of either study group.
Results from this phase 3 randomized, placebo-controlled study provide further evidence of the efficacy and tolerability of ospemifene 60 mg/day for treating symptoms of VVA associated with menopause.10,21 Twelve weeks of oral treatment with ospemifene 60 mg/day induced statistically significant beneficial changes in the four coprimary efficacy endpoints: parabasal cells decreased by 40.2% and superficial cells increased by 12.3% (compared with a 0% reduction and a 1.7% increase in the placebo group; P < 0.0001), vaginal pH decreased by −0.94 (compared with −0.07 in the placebo group; P < 0.0001), and the self-reported MBS severity score of dyspareunia decreased by −1.5 (compared with −1.2 in the placebo group; P = 0.0001).
Notably, in the secondary endpoint analysis, improvements in vaginal pH, parabasal cells, and superficial cells were observed at the earliest point assessed, 4 weeks after initiation of treatment. The improvements in the vaginal epithelium and vaginal pH related to VVA were reflected in the superior efficacy of ospemifene compared with placebo in reducing vaginal pain associated with sexual intercourse. In addition, there was a reduction in nonhormonal lubricant use in the ospemifene group compared with the placebo group, despite no decrease in the frequency of sexual activity in either of the two groups. Many women in the postmenopausal population use over-the-counter lubricants but continue to have painful intercourse. The beneficial impact of ospemifene on sexual health observed in this study where women concomitantly used over-the-counter lubricants is likely to extend to women in the general population. Further research and analyses are needed to evaluate whether sexual function and quality-of-life measures in this population improve as well.
Consistent with previously reported trials, ospemifene was shown to be safe and well tolerated. Hot flushes occurred more frequently in the ospemifene group, as was the case in another clinical trial.10 However, discontinuation rates for hot flushes were low and similar in both groups, as only one participant in the ospemifene group and one participant in the placebo group discontinued because of hot flushes.
Endometrial safety has been an important consideration during the clinical development of new SERMs, especially when they are being considered for use in the management of a chronic condition such as VVA.15 In this study, no clinically meaningful estrogenic effects on the endometrium were seen. Endometrial thickness, as assessed by TVU, increased slightly. Based on endometrial biopsies, no cases of endometrial hyperplasia or carcinoma were reported. This study adds to evidence from animal data and clinical trials indicating that ospemifene does not have clinically relevant effects on the uterus or endometrium.21-25
In addition to this study, two other phase 3 studies have evaluated the tissue selectivity of ospemifene. One study—a 12-week initial efficacy and safety study with an additional 40 weeks of long-term safety (total 52 wk)10,19—demonstrated no clinically meaningful endometrial changes with long-term treatment with oral ospemifene 60 mg/day in 180 postmenopausal women with VVA.19 A second study of 52 weeks similarly demonstrated minimal estrogenic effects on the endometrium.20
The present study demonstrated the efficacy and safety of ospemifene in the largest population of postmenopausal women with dyspareunia reported to date. Ospemifene provides a needed oral therapeutic option for the large population of postmenopausal women who experience moderate to severe vaginal dyspareunia caused by the chronic hypoestrogenic state of menopause that does not resolve spontaneously and often worsens with age. Because women are living longer5 and many continue to be sexually active past menopause,26 a well-tolerated and effective treatment could potentially benefit a great number of postmenopausal women and their partners.
Although the population may not be completely representative of that seen in clinical practice since it focused on only one of the several common symptoms experienced by a majority of women with postmenopausal VVA, the design of the study, as well as the large sample size, did make it appropriate for assessing the safety and efficacy of ospemifene.
Ospemifene 60 mg/day, compared with placebo, effectively reduces the physiological signs of VVA and dyspareunia in postmenopausal women while not inducing significant estrogenic effects on endometrial tissue or clinically important AEs. In this phase 3 study, ospemifene 60 mg/day demonstrated significant improvements relative to placebo for all coprimary endpoints in both the ITT population and the PP population and was generally well tolerated. Furthermore, ospemifene 60 mg/day demonstrated significant efficacy in the presence of lubricant use, with numerically less lubricant use observed in the treated cohort compared with participants given placebo. There are no currently available SERMs with a tissue-selective profile conducive for the treatment of postmenopausal VVA (ie, estrogenic action on the vaginal epithelium concurrent with minimal effects on the endometrium and neutral effects on the breast). Based on the results of this trial and previously reported trials, ospemifene has the potential to be the first oral alternative to vaginal estrogens for the treatment of dyspareunia associated with VVA.
Ospemifene study group (clinical centers—principal investigators): Adams Patterson Gynecology & Obstetrics, Memphis, TN — Judith Williams; Advanced Clinical Research, West Jordan, UT—Judith Kirstein; Quality of Life Medical and Research Center, Tucson, AZ—Walter Patton; Center for Women’s Health, Plainsboro, NJ—Scott Eder; Essential Women’s Associates, Henderson, NV—Edmond Pack; Centennial Hills OB-Gyn Associates, North Las Vegas, NV—Nader Abdelsayed; Affiliated Clinical Research, Inc, Las Vegas, NV—R. Garn Mabey Jr; Altus Research, Lake Worth, FL—Samuel Lederman; American Health Network of Indiana, LLC, Franklin, Greenfield, and Avon, IN—Jill Beavins and Joseph R. Schnecker; Atlantic Institute of Clinical Research, Daytona Beach, FL—Thomas Stavoy; BlueGrass Clinical Research, Inc., Louisville, KY—Arthur Donovan; Advanced Research Associates, Corpus Christi, TX—Charles Eubank; Center for Fertility and Women’s Health, PC, New Britain, CT—Anthony Luciano; Clinical Research of South Florida, Coral Gables, FL-Jeffrey Rosen; Clinical Trials of Virginia, Inc., Richmond, VA—Bruce Johnson; ClinSite, LLC, Ann Arbor, MI—Gayle Moyer and Donna Hrozenzik; Columbus Center for Women’s Health Research, Inc., Columbus, OH—David J. Portman; Comprehensive Clinical Trials, LLC, West Palm Beach, FL—Ronald Ackerman; Eastern Carolina Women’s Center, New Bern, NC—Jeffrey Michelson; Fellows Research Alliance, Inc, Savannah, GA and Hilton Head Isle, SC—Debra Walland; Holston Medical Group, P.C. (Seasons), Bristol, TN—Dennis Samuel; Holzer Clinic, Inc, Gallipolis, OH—James Strafford; Horizons Clinical Research Center, LLC, Denver, CO—Theodore Cooper; Jackson Clinic, PA, Jackson, TN—Stephen Hammond; Lyndhurst Clinical Research/Lyndhurst Gynecologic Associates, Winston-Salem, NC—Robert Parker Jr; Medical Center for Clinical Research, San Diego, CA—William Koltun; Medisphere Medical Research Center, LLC, Evansville, IN—Steven Elliott; Meridien Research, Inc, St. Petersburg, FL—Gigi Lefebvre; Mount Vernon Clinical Research, LLC, Sandy Springs, GA—Stephen Blank; OB/GYN Specialist of the Palm Beaches, P.A., West Palm Beach, FL—John Burigo; Philadelphia Clinical Research, LLC, Philadelphia, PA—Larry Seidman; Phoenix OB-GYN Associates, Moorestown, NJ—Bruce Levine; Puget Sound Osteoporosis Center, Seattle, WA—Susan Nattrass; Radiant Research, Atlanta, GA, Chicago, IL, Cincinnati, OH, and San Antonio, TX—Sidney Funk, Phyllis Marx, Michael Noss, and William Jennings; Springfield Clinic, LLP, Springfield, IL—Randolph Roller; Suncoast Clinical Research, Inc, New Port Richey, FL—Robert Smith; Tidewater Clinical Research, Inc, Virginia Beach, VA—Franklin Morgan Jr.; Valley Women’s Clinic, PLLC, Renton, WA—James Rice; Virginia Women’s Center, Richmond, VA—Peter Zedler; Visions Clinical Research, Boynton Beach, FL—Keith Aqua; Wake Research Associates, LLC, Raleigh, NC—Pouru Bhiwandi; Westlake Medical Research, Inc, Westlake Village, CA—Edward Portnoy; Woman’s Clinic, LLP, Boise, ID—Kerry B. Lowder; Women’s Clinic of Lincoln, PC, Lincoln, NE—Stephen Swanson; Hudson Valley Urology, PC, Poughkeepsie, NY—Evan Goldfischer; Clinical Trials of America, Inc, Eugene, OR—Richard Beyerlein; ObGYN Associates, Richmond, VA—Boyden B. Clary III; Hawthorne Medical Research, Winston-Salem, NC—Richard Hedrick Jr; Jacksonville Center for Clinical Research, LTD, Jacksonville, FL—Neil Sager; COR Clinical Research LLC, Oklahoma City, OK—Clinton Corder; IGO Medical Group, San Diego, CA—Philip Young; Lyndhurst Clinical Research, Kerersville, NC—Melvin Seid; Clinical Physiology Associates, Inc, Ft. Myers, FL—Mary Yankaskas; Downtown Women’s Health Care, Denver, CO—Arthur Waldbaum; Nature Coast Clinical Research, LLC, Crystal River, FL—Scott Redrick; North Spokane Women’s Clinic, Spokane, WA—Ronald Hardy.
We thank the study staff at each site and all the women who participated in this study.
Technical and editorial assistance was provided by Diane Kwiatkoski, PhD, and Rebecca A. Bachmann, PhD (not related to author Gloria A. Bachmann, MD) of Quintiles.