Faculty Bibliography

OBJECTIVE: The aim of this study was to report the gynecologic safety findings from the Generations trial, a phase 3 study of the selective estrogen receptor modulator (SERM), arzoxifene. METHODS: A predefined objective of the trial was to evaluate the effects of arzoxifene on the genital tract. Gynecologic examinations were performed yearly, and further gynecologic assessment, including endometrial biopsy, was performed in a predefined subset of women and in those who developed vaginal bleeding. RESULTS: Overall, 9,354 women were randomized (4,678 to placebo, 4,676 to arzoxifene 20 mg/d). There were 13 adjudicated cases of endometrial cancer (placebo, 4 cases; arzoxifene, 9 cases: P = 0.165) and 6 cases of endometrial hyperplasia (placebo, 2 cases; arzoxifene, 4 cases). Endometrial thickness, assessed at 24- and 36-month transvaginal ultrasounds in a subset of women, increased slightly in women assigned to arzoxifene compared with baseline and women in placebo. At the last measurement, 3 (1.7%) women assigned to placebo and 21 (10.2%) assigned to arzoxifene had an endometrial thickness greater than 5 mm (P < 0.001 for difference between treatment groups). Endometrial polyps were more common in women treated with arzoxifene (n = 37) than in women treated with placebo (n = 18; P < 0.05). Vulvular and vaginal inflammation, including mycotic infections, and vaginal discharge were reported more frequently in women treated with arzoxifene than in women treated with placebo, as were urinary tract infections. CONCLUSIONS: Gynecologic events were generally more common in women treated with arzoxifene than in women treated with placebo. The gynecologic effects of arzoxifene seem to differ from those of raloxifene, although both SERMs have a benzothiophene structure. Although all SERMs influence cells through the estrogen receptor, they need to be evaluated independently in terms of their effects on various tissues, including the genital tract.

Objective: Currently, only estrogen-based preparations have received regulatory approval for prescription treatment of vulvovaginal atrophy (VVA) in postmenopausal women; therefore, exploring alternative treatments for WA is imperative. Ospemifene, a novel, selective estrogen receptor modulator (SERM), is unique compared with other SERMs due to its estrogenic activity in the vaginal epithelium and is under investigation in postmenopausal women with VVA symptoms. This study assessed the efficacy, safety, and tolerability of ospemifene 60 mg/d in the treatment of WA symptoms in postmenopausal women. Design: This 12-wk, randomized, double-blind, placebo-controlled, parallel-group study included 919 subjects. Two study populations were analyzed: intent-to-treat (ITT), the primary analysis, and per protocol (PP). Women (4080 years of age) were assigned to one of 2 VVA symptom strata (vaginal dryness or dyspareunia) based on their self-reported most bothersome WA symptom (MBS);and within each stratum were randomized 1:1 to receive either ospemifene 60 mg/d or placebo. Data from each stratum were independently analyzed. All subjects were provided with a nonhormonal vaginal lubricant to be used as needed (PRN). For each stratum, co-primary efficacy endpoints were measured for change from baseline to Wk 12 (LOCF) in: vaginal pH, % superficial cells and % parabasal cells in the maturation index, and the severity of the MBS. This abstract reports the dyspareunia stratum results. Results: At Wk 12 ospemifene demonstrated significant efficacy vs placebo for each co-primary endpoint. Changes in the ITT population for vaginal pH (LS mean),% of superficial cells (median), and % of parabasal cells (LS mean) all were p<0.0001 (Table 1); a significant effect was observed as early as 4 wks. Significant mean improvement was also observed for the MBS, dyspareunia (p=0.0001) at Wk 12. A higher percentage of subjects treated with ospemifene reported no, or mild dyspareunia and self-reported symptom severity that improved by 2-3 points on a 4-point Likert scale in 52.8% of ospemifene vs 38.8% of placebo subjects. The PP analysis also demonstrated statistically significant findings for each co-primary endpoint, confirming ll'i analysis. The % of subjects with at least 1 adverse event (in the ITT population) at Wk 12 for both strata combined is summarized in Table 2. Discontinuation rates were similar in the ospemifene and placebo groups, 10.2% vs 11.6%, respectively. Endometrial histology assessments showed no cases of hyperplasia and 2 (1.0%) cases of active proliferation in the ospemifene group vs 0% in the placebo group. Vaginal bleeding was reported in 2 (0.4%) and 4 (0.9%) subjects in the ospemifene and placebo groups, respectively. One subject in the ospemifene group experienced a deep vein thrombosis following an 8-hour car ride and was discontinued from the study. Conclusion: In postmenopausal women with self-reported MBS of dyspareunia, this unique study demonstrated that, despite the PRN use of lubricants, treatment with ospemifene 60 mg/d demonstrated clinically and statistically significant efficacy and was well tolerated. (Table Presented)

Objective: Ospemifene, a novel, selective estrogen receptor modulator (SERM) that exerts estrogenic, pharmacologic activity in the vaginal epithelium, is presently being studied for treatment of symptoms of vulvovaginal atrophy (WA) in postmenopausal women. This study assessed the efficacy, safety and tolerability of ospemifene 60 mg/d in the treatment of WA symptoms. Design: A 12-wk, 1:1 randomized, double-blind, placebo-controlled, parallel-group study enrolling 919 postmenopausal women 40 to 80 years of age with WA in two strata based on their self-reported most bothersome symptom (MBS) of vaginal dryness or vaginal pain (dyspareunia). Two study populations were analyzed: the intent-to-treat (TTT) (primary analysis) and per protocol (PP). Subjects in each stratum were randomized to receive 60 mg/d ospemifene or placebo and were provided with a nonhormonal vaginal lubricant to use as needed (PRN). For each stratum, changes from baseline to Wk 12 (LOCF) for the four co-primary endpoints were assessed: vaginal pH, percentages of superficial cells and parabasal cells in the maturation index and the severity of the MBS. This abstract reports the Dryness Stratum results. Results: In the TTT analysis, at Wk 12 ospemifene demonstrated significant efficacy vs placebo for 3 of 4 co-primary endpoints from baseline. Significant mean changes from baseline to Wk 12 for vaginal pH and percentages of superficial (LS mean) and parabasal cells (Median) were evident (Table 1). Significant improvement was observed as early as 4 wks. Improved mean change was observed for the MBS vaginal dryness at Wk 12, which approached statistical significance (P=0.0803). A higher % of subjects treated with ospemifene reported no vaginal dryness, and the subjects' self-reported symptom severity, which was assessed on a 4-point scale improved by 2 to 3 points in 46.3% ospemifene vs 34.4% placebo subjects. The PP analysis showed statistically significant improvement for all 4 co-primary endpoints (pH, % superficial and % parabasal cells, all P<0.0001 and vaginal dryness, P=0.0143) and similar improvements in dryness severity. The main difference between the ITT and the PP populations was in study drug compliance, which was higher in the PP population. The numbers of subjects with >1 adverse event (AE) at Wk 12 in both strata combined is summarized in Table 2. Discontinuation rates were similar in the ospemifene (10.2%) and placebo (11.6%) groups. Endometrial histology assessments showed no cases of hyperplasia and 2 (1.0%) cases of active proliferation in the ospemifene group vs 0% in the placebo group. Vaginal bleeding was reported in 2 (0.4%) and 4 (0.9%) subjects in the ospemifene and placebo groups, respectively; 1 subject on ospemifene experienced deep vein thrombosis was discontinued from the study. There were no cases of myocardial infarction, breast cancer or death. Conclusion: In postmenopausal women with the self-reported MBS of vaginal dryness, these data demonstrate that treatment with ospemifene 60 mg/d provides clinically and statistically significant efficacy and was well tolerated. With greater improvement in symptom severity scale changes and in markers of vaginal health, this novel SERM may prove to be the first non-estrogen to effectively treat the symptoms of WA. (Table Presented)

The ascertainment of diagnostic medical information through most of the 20th Century relied on techniques like inspection, palpation, and auscultation as noninvasive ways to obtain information about internal anatomy, both normal and abnormal, as well as physiology which when not normal we call "pathology". Sensitivity of imaging modalities has become so refined that we are seeing structures either never before appreciated or sometimes appreciated but not with such precision. This lecture will deal with my observations of this issue in gynecology using transvaginal ultrasound (TV U/S), although the problems associated with increased sensitivity of imaging exist in many other clinical fields as well. My thesis is that as we see more and more detail the response to what we see has often been based on old and thus preconceived notions, and not on newer well-organized and well-designed studies to determine the prevalence of particular findings and their clinical significance. Information obtained with increasingly refined technology cannot simply be handled according to old established principles. Newer studies must be made before clinical recommendations can be made. We must be careful not to over interpret such findings that may be much more common and less ominous that previously believed. The vaginal probe has revolutionized gynecology. Higher frequency probes in close proximity to the structure being studied have created a degree of image magnification that is as if we are doing ultrasound through a low power microscope (sonomicroscopy). We are seeing things with ultrasound that you cannot see with the naked eye. Furthermore, the ease of performing TV U/S, and its increasingly routine use for surveillance, has resulted, in my opinion, in an explosion of incidental findings that are much more common and more innocuous than previously realized. There are many examples in gynecology but I will concentrate on two that are very common and relevant to menopausal patients. These are 1) Simple ovarian cystic structures and 2) Incidental endometrial "thickening" in non bleeding postmenopausal women. Forty years ago the dictum that a palpable postmenopausal ovary was abnormal was put forth by Dr. Hugh Barber. When ultrasound (then transabdominal ultrasound) began to see postmenopausal cystic structures these were promptly removed as potentially malignant. 25 years ago the first small observational study began to suggest that unilocular unilateral cysts were invariably benign and could be managed expectantly. Abundant data from prospective screening studies have confirmed that observation. They have also shown the incidence of unilocular cystic adnexal structures in postmenopausal women can be as high as 18%. The field has "matured" to where almost all clinicians will now counsel such patients for conservative management. Twenty years ago the first reports of a thin distinct endometrial echo in postmenopausal patients with bleeding surfaced. Abundant prospective data from multicentered trials showed that the incidence of malignancy in postmenopausal patients with bleeding when the endometrial echo measures < 4mm on transvaginal ultrasound is 1/917. In 2009 ACOG stated "when transvaginal ultrasound is performed for patients with postmenopausal bleeding and an endometrial thickness <4mm is found endometrial sampling is not required". Unfortunately most clinicians who understood that a thin distinct endometrial echo in such patients excludes malignancy, have still inappropriately assumed that a thick echo in non-bleeding patients is abnormal and requires intervention. The prevalence of non thin endometrial echo on TV U/S in asymptomatic postmenopausal women is 10-17%. Furthermorel3% of asymptomatic postmenopausal women will have an endometrial polyp on TV U/S. The risk of malignancy is such imaged structures, absent bleeding, is less than 1 in 250 yet the complication rate from operative hysteroscopy in this elderly group is as high as 3.6%. The increasing sensitivity of the vaginal probe can and should be a wonderful tool for the gynecologist. However studies on the prevalence and significance of the things one can see on such imaging must be carried out if it is to be incorporated appropriately. Remember the credo -"above all else do no harm".

Postmenopausal bleeding is "endometrial cancer until proven otherwise". The incidence of cancer is reported as 1-14% although most studies are in the 3-7% range. The majority will actually bleed from inactive atrophic endometrium. In the early 1990's pilot studies began to suggest that a thin distinct endometrial echo <4-5 mm in postmenopausal patients with bleeding was associated with a lack of significant tissue and effectively excluded endometrial carcinoma. Since then several large prospective multinational trials in Europe have corroborated this. Taken together it appears that in patients with postmenopausal bleeding a thin distinct endometrial echo <4mm on transvaginal has a risk of endometrial cancer of 1 in 917. It is crucial that the examiner appreciate that not all uteri lend themselves to a meaningful ultrasound examination. Axial uterus, previous surgery, coexisting myomas, marked obesity, adenomyosis will result in a significant minority of such patients not yielding a reliable endometrial echo visualized with transvaginal sonography. In such cases saline infusion sonohysterography (SIS) may be helpful. SIS should be thought of as a subset of TV/US in cases in which the endometrium is not thin or not well visualized. Blind endometrial biopsy with suction piston biopsy instruments has been a standard approach in spite of a lack of appropriate validation especially in light of the fact that endometrial pathology is not always global. Transvaginal ultrasound can and should be the first approach to the postmenopausal patient with bleeding

Selective estrogen receptor modulators (SERMs) can act as estrogen agonists or estrogen antagonist depending on the target tissue. Tamoxifen was the first clinically available SERM and is highly effective for the prevention and treatment of breast cancer. Tamoxifen however has estrogen agonistic activity in the uterus and has been associated with an increased risk of endometrial hyperplasia and malignancy. Thus endometrial safety has been an important consideration in the development of all SERMs since then. Raloxifene, which is currently approved for prevention and treatment of postmenopausal osteoporosis, and for the prevention of breast cancer, does not result in endometrial hyperplasia or cancer. It does result in a slight increase in endometrial polyps. Lasofoxifene which has been shown to reduce fracture risk and decrease the incidence of breast cancer did not increase endometrial cancers or endometrial hyperplasia but did result in an increased incidence in vaginal bleeding, endometrial polyps ( all of which were inactive and atrophic) and endometrial "thickening" on transvaginal ultrasound. Ospemefene has been shown to have beneficial effects on the vaginal epithelium but does not cause endometrial cancer or hyperplasia. Bazodoxifene is effective in preventing and treating postmenopausal osteoporosis and does not cause endometrial thickening on ultrasound nor any hyperplasia or cancer. Arzoxifene was evaluated in a phase III trials for treating osteoporosis but its clinical development program was suspended. Further investigation of newer SERMs is warranted to more clearly define the endometrial safety of each of these agents

Introduction. Currently, only estrogen-based prescription preparations have received regulatory- authorization for the prescription treatment of vulvar and vaginal atrophy (VVA) in postmenopausal women. Therefore, clinical development of tissue-selective therapies to treat VVA which lack the undesired systemic effects of estrogen is warranted. Ospemifene, a selective estrogen receptor modulator (SERM), is differentiated from other SERMs by its estrogenic action in the vaginal epithelium, not on the endometrium, and is under investigation for use in postmenopausal women with VVA symptoms. This study evaluated the safety, efficacy, and estrogenic effects on the endometrium of ospemifene for up to 1 year; this abstract reports data for overall safety. Methods. A 52-wk, 6:1 randomized, double-blind, placebo-controlled, parallel-group study was conducted comparing 60 mg/d of oral ospemifene (n=363) with placebo (n=63) in postmenopausal women with VVA and an intact uterus. Clinical safety variables were assessed at screening, randomization and wks 12, 26, 39, 52 and 56. Results. A total of 22 (5.2%) subjects experienced >=1 SAEs. A numerically higher proportion of serious AEs (SAEs) were reported in the placebo group (6.5%) vs the ospemifene group (4.9%). In the ospemifene group, 1 (0.3%) subject had deep vein thrombosis, which resolved. No cases of pulmonary embolus, retinal vein thrombosis, or myocardial infarction were reported, and no subjects developed breast or endometrial cancer. Similar proportions of subjects (placebo, 75.8%; ospemifene, 84.6%) reported >1 treatment-emergent adverse event (AE); most were considered mild or moderate. The most commonly reported AE in the ospemifene group was hot flush (6.5% and 12.6%) for placebo (mean age 62.9 years) and ospemifene (mean age 61.7 years), respectively. The discontinuation rate for subjects in the ospemifene group who experienced hot flushes was approximately 2%.. Conclusion. Ospemifene 60 mg/d was found to be safe and well tolerated, in addition to efficacious, when used for the treatment of VVA in postmenopausal women with an intact uterus for up to 1 year