Estradiol Valerate/Dienogest: A Novel Combined Oral Contraceptive

Clinical Therapeutics/Volume 34, Number 1, 2012 New Drug Estradiol Valerate/Dienogest: A Novel Combined Oral Contraceptive Laura M. Borgelt, PharmD1; and Chad W. Martell, PharmD2 1 Departments of Clinical Pharmacy and Family Medicine, Anschutz Medical Campus, University of Colorado, Aurora, Colorado; and 2Department of Pharmacy Practice, Rueckert-Hartman College for Health Professions, Regis University, Denver, Colorado ABSTRACT Background: Estradiol valerate/dienogest (E2V/ DNG) is a combined oral contraceptive (COC) with 2 new hormonal entities and a unique 4-phasic dosing regimen indicated for women to prevent pregnancy. Objective: The purpose of this article is to review the pharmacology, pharmacokinetics, clinical efficacy, tolerability, and cost of E2V/DNG. Methods: MEDLINE (1966 –June 2011) and EMBASE (1966 –June 2011) were searched for original research and review articles published in the English language using the terms Natazia or Qlaira or estradiol valerate and dienogest. The reference lists of identified articles were reviewed for additional pertinent publications. Abstracts from the 2005 to 2011 American Society of Reproductive Medicine and American College of Obstetricians and Gynecologists meetings were searched using the same terms. Results: The search provided 56 articles that addressed the pharmacology, pharmacokinetics, pharmacodynamics, clinical efficacy, and tolerability of E2V/DNG in women of reproductive age. Articles reporting efficacy or tolerability in the setting of menopause were excluded. The initial efficacy of E2V/DNG on ovulation inhibition was investigated in 2 prospective, randomized, open-label, Phase II dose-finding studies. The dose that was approved by the Food and Drug Administration resulted in 3.13% of women ovulating in the second cycle of treatment (90% CI, 0.2%– 6.05%). Rate of pregnancy prevention with this agent was reported with a Pearl Index ranging from 0.73 to 1.27 (unadjusted) to 0.34 to 0.72 (adjusted for method failure only). The mean duration of withdrawal bleeding was 4.3 days (range, 4.0 – 4.6 days) among 2266 women receiving 13 treatment cycles. Adverse events reported in ⬎1% of patients included ab- January 2012 dominal pain, acne, breast pain, dysmenorrhea, emotional lability, headache, nausea, and weight increase. Conclusions: Estradiol valerate/dienogest is a new contraceptive formulation. It offers efficacy, tolerability, and an acceptable safety profile with a potentially better bleeding pattern than levonorgestrel-containing COCs. This COC may be especially useful for older women of reproductive age who are adherent to therapy and looking for shorter and/or lighter menstrual cycles. Studies will need to be performed to determine whether clinically significant differences in outcomes exist among E2V/DNG and other available COCs. (Clin Ther. 2012;34:37–55) © 2012 Elsevier HS Journals, Inc. All rights reserved. Key words: contraception, combined oral contraceptives, dienogest, estradiol valerate, oral contraceptives. INTRODUCTION Combined oral contraceptives (COCs) contain both an estrogen and progestin component. Significant developments and changes in COCs have occurred since their introduction in the 1960s. The first combined contraceptive pill marketed in the United States in 1961 contained 5 mg of norethynodrel and 75␮g of mestranol.*1 Although a higher dose of this agent containing 9.85 mg of norethynodrel and 150␮g of mestranol had been approved in 1960, it was never marketed as a contraceptive.2 These high doses of hormones were used because they were found to be effective and, at that time, studies Accepted for publication November 2, 2011. doi:10.1016/j.clinthera.2011.11.006 0149-2918/$ - see front matter © 2012 Elsevier HS Journals, Inc. All rights reserved. *Trademark: Enovid™(G. D. Searle & Company, Skokie, Illinois). 37 Clinical Therapeutics were not performed to identify the lowest effective dose. Since those first pills, the doses of sex steroids in COCs have decreased significantly, which has allowed for increased safety and decreased adverse effects. Research efforts since the 1960s have focused on the effect of estrogen and progestin for effective and safe contraception. Three estrogenic compounds are used in the oral contraceptives (OCs) available in the United States. Mestranol is a prodrug that must be metabolized by the liver into ethinyl estradiol (EE) to have biologic activity. This component was used frequently in the early COCs but is now found only in a few COCs at a 50-␮g dose. This mestranol dose is equivalent to approximately 35 to 40 ␮g of EE.3 Most COCs available today contain EE as the estrogen component. Ethinyl estradiol has been synthesized since 1938 because the natural estradiol is poorly absorbed when taken orally and is rapidly inactivated by the liver.1 The substitution at C17 with an ethinyl group on the estradiol component is much more resistant to degradation.4 This creates a much more potent and longer-acting compound. It allows for once-daily dosing but also has a greater effect on metabolic parameters compared with estradiol.4 Doses of EE in COCs in the United States vary from 20 to 50 ␮g daily. Recently, a new estrogen component, estradiol valerate (E2V), has been introduced into a COC in the United States. It is immediately cleaved to estradiol so the circulating molecule reaching the estrogen receptors is the natural 17␤-estradiol.5 Although estradiol is effective in helping to prevent pregnancy, the 17␤-estradiol in its natural form causes poor cycle control.6 When E2V is combined with dienogest (DNG) in multiphasic regimens, it solves the problem of poor cycle control observed with previous 17␤-estradiol COCs.6,7 The progestin component provides most of the contraceptive activity of the COC. Nine different progestins have been used in the COCs sold in the United States, each with a different potency and different metabolic effects. The progestins are typically categorized into “generations” based on when they were introduced into the United States. The first-generation progestins are norethindrone, norethindrone acetate, and ethynodiol diacetate. Because the dose of these agents was reduced over time, some women began experiencing more unscheduled bleeding and spotting; secondgeneration progestins were developed in response to these unwanted effects.3 The second-generation progestins are norgestrel and levonorgestrel (LNG). These 38 compounds are more potent and have longer half-lives than the first-generation progestins.3 Pills containing these progestins have more androgenic activity, which may be helpful for libido but detrimental for women with hirsutism, acne, or dyslipidemia.3 Third-generation progestins, desogestrel and norgestimate, were designed to maintain increased progestational activity but reduce androgenic activity. With less androgenic activity, the estrogen component can more fully express itself metabolically.3 This may be effective for acne, but the increased expression of estrogen may lead to a higher risk of thromboembolic events, especially in COCs containing EE.3,8 The fourth-generation progestins, drospirenone and DNG, have been designed to specifically bind to the progesterone receptor without any interaction with other steroid receptors.8 Several innovations have been introduced over the years in COC formulations and packaging. Formulations may vary by the type and amount of hormones, the patterns of those amounts throughout the cycle, and the number of active pills in the packet. Monophasic formulations have active pills that contain the same amount of estrogen and progestin. Multiphasic formulations have estrogen and progestin amounts that vary. Biphasic formulations have 2 different combinations of estrogen and progestin, and triphasic formulations have 3 different combinations. Recently, a 4-phasic formulation has been introduced with an estrogen step-down and progestin step-up sequence.9 Most COC packages contain 21 active (containing hormone) pills with 7 placebo (inert) pills; these packages are known as 21/7 regimens. During the placebo week, a withdrawal bleed occurs. Early in COC development, this was valid because at that time rapid pregnancy tests were not available and it reassured women they were not pregnant. As doses of estrogen and progestin have decreased in COCs, serum levels decrease low enough for endometrial sloughing to begin within 2 to 3 days of the last active pill.3 Decreasing the number of placebo pills in low-dose formulations is necessary to prevent recruitment of follicles.10,11 Therefore, several formulations with shortened hormone-free intervals are now available in 24/4 or 26/2 regimens. As COCs have evolved, it has become clear that individual women have individual needs for contraception. Approximately 38 million women (98% of 43 million fertile, sexually active women) in the United States are practicing contraception.12 Twenty-eight percent of them are choosing a COC as their primary method.12 The pill is Volume 34 Number 1 L.M. Borgelt and C.W. Martell O CH3 O H H H HO Figure 2. Estradiol valerate.9 Figure 1. Dosing schedule for the four-phasic estradiol valerate/dienogest combined oral contraceptive. the most commonly used method by women in their 20s, women who are cohabitating, women with no children, and women with at least a college degree.12 With so many women choosing this method of contraception, an agent that provides high efficacy with an excellent safety profile and convenience is ideal. A novel 4-phase COC that contains E2V and DNG† was approved by the Food and Drug Administration in May 2010 for use by women to prevent pregnancy.9 It is formulated with an estrogen step-down and progestin step-up sequence. Specifically, the 28-tablet pack contains 2 tablets containing 3 mg of E2V, 5 tablets containing 2 mg of E2V and 2 mg of DNG, 17 tablets containing 2 mg of E2V and 3 mg of DNG, 2 tablets containing 1 mg of E2V, and 2 inert tablets (Figure 1).9 This article reviews the pharmacology, pharmacokinetics, clinical efficacy, tolerability, and cost of E2V/DNG. METHODS and further analyzed. Animal studies were also considered for review of the pharmacology and pharmacokinetics of the medication. Articles reporting clinical efficacy or tolerability in the setting of menopause treatment were excluded from the review. The reference lists of identified articles were reviewed for additional pertinent publications. Abstracts from the 2005 to 2011 American Society of Reproductive Medicine and American College of Obstetricians and Gynecologists meetings were searched using similar terms. RESULTS Clinical Pharmacology Estradiol valerate is chemically described as estra1,2,5(10)-triene-3,17-diol(17␤)-,17-pentanoate, and its empirical formula is C23H32O3.91 Dienogest is chemically described as (17␣)-17-hydroxy-3-oxo-19norpregna-4,9-diene-21-nitrile, and its empirical formula is C20H25NO2.9 The chemical structure of each of these components is shown in Figure 2 and Figure 3. Mechanism of Action Although estrogen and progestin each have a distinctive role in preventing pregnancy, their primary ef- MEDLINE (1966 –June 2011) and EMBASE (1966 – June 2011) were searched for original research and review articles published in the English language using the terms Natazia or Qlaira or estradiol valerate and dienogest. The search provided 56 articles that were considered for inclusion in the review. Articles that addressed the pharmacology, pharmacokinetics, pharmacodynamics, clinical efficacy, and tolerability in women of reproductive age were selected for inclusion † Trademark: Natazia™ (Bayer Healthcare Pharmaceuticals Inc., Wayne, New Jersey). January 2012 Figure 3. Dienogest.9 39 Clinical Therapeutics fect is to suppress gonadotropins and ovulation. The estrogen component prevents an increase in folliclestimulating hormone to prevent a dominant follicle from emerging, whereas the progestin inhibits luteinizing hormone to prevent ovulation.13 In addition, COCs suppress ovulation by providing negative feedback to the hypothalamic-pituitary system through decreased gonadotropin-releasing hormone pulsatility and decreased pituitary responsiveness to gonadotropin-releasing hormone stimulation.3 The estrogen also provides endometrial proliferation to minimize unwanted breakthrough spotting or bleeding and potentiates the action of the progestin component (which allows for lower doses of progestational agents to be used). The progestin opposes the mitotic action of estrogen, causing a stable decidualized endometrium that is not receptive to implantation.13,14 In addition, progestins cause a thickening of the cervical mucus to inhibit sperm transport. The combination of these mechanisms serves to provide contraceptive efficacy. Perfect use of COCs has been shown to have a failure rate of 0.3% in the first year of use, and typical use has a failure rate of 8.7%.12 Pharmacokinetics A phase 1, open-label study of 15 healthy women aged 18 to 50 years was conducted to evaluated the pharmacokinetics of E2V/DNG.15 Participants were nonsmokers with a body mass index (BMI; calculated as weight in kilograms divided by height in meters squared) of 18 to 26 and a follicle-stimulating hormone level of 10 mIU/mL or less. Women underwent a screening period of up to 3 weeks before receiving treatment, and treatment was given for 4 weeks. Follow-up occurred for the next 3 weeks, and blood samples were collected for 10 weeks. The treatment regimen included the 4-phase dosing regimen for 26 days of active treatment. Treatment was given for 4 weeks, and follow-up occurred for the next 3 weeks. Blood samples were collected for 10 weeks. The following section describing absorption, distribution, metabolism, and elimination of E2V/DNG compiled data from this study, other pharmacokinetic studies, product information, and several detailed reviews of the drug.9,14 –18 Geometric means or %CV values are reported unless otherwise stated. intestinal mucosa or first-pass metabolism in the liver.18 Approximately 3% of oral E2V is bioavailable as estradiol. One milligram of E2V contains 0.76 mg of 17␤-estradiol.15,18,19 Serum estradiol Cmax values of 0.0706 ng/mL (%CV, 27.8%) on day 1 and 0.0660 ng/mL (%CV, 39.9%) on day 24 were achieved at a median time of 6 hours on day 1 and 3 hours on day 24 (Tmax, 1.5–12 hours).15 The mean AUC0 –24 of estradiol was similar on days 1 and 24 with values of 1.246 (%CV, 29.2%) and 1.239 (%CV, 39.9%) ng · h/mL, respectively. Although not clinically significant, concomitant food intake resulted in a 23% increase in Cmax, whereas the AUC did not change.9,17 Serum estrone and estrone sulfate values of 0.416 ng/mL (%CV, 54.1%) at a median time of 4 hours (range, 3–12 hours) and 16.384 ng/mL (%CV, 53.5%) at a median time of 3 hours (range, 1.5–10 hours) were achieved on day 1. Maximum serum estrone and estrone sulfate concentrations were 0.444 ng/mL (%CV, 44.9%) at a median time of 4 hours (range, 3–12 hours) and 13.478 ng/mL (%CV, 54.3%) at a median time of 3 hours (range, 1.5–12 hours) on day 24, respectively. The mean AUC0 –24 of estrone was 5.750 ng · h/mL (%CV, 52.9%) on day 1 and 6.814 ng · h/mL (%CV, 52.1%) on day 24. The mean AUC0 –24 of estrone sulfate was 177.489 ng · h/mL (%CV, 56.5%) on day 1 and 163.820 ng · h/mL (%CV, 56.0%) on day 24. The overall estrone:estradiol serum ratio was approximately 5:1.15 Minimum mean serum concentrations (ie, trough concentrations) were stable throughout treatment, with values ranging from 0.0336 to 0.0647 ng/mL.15 After oral administration of DNG, pharmacokinetics are dose proportional, within the dosing range of 1 to 8 mg.9,17 Dienogest is nearly completely absorbed with a bioavailability of approximately 91%.9,20 Food intake decreased the Cmax by 28% but did not change the AUC and is not clinically relevant.9,17 On day 24, the Cmax of DNG increased to 82.9 ng/mL (%CV, 25.7%) within 1.5 hours (range, 1–2 hours).15 Minimum concentrations ranged from 6.8 to 15.1 ng/mL. The mean DNG concentration at steady state on day 24 was 33.7 ng/mL (%CV, 22.5%). The AUC24 value on day 24 was 809 ng · h/mL; the mean accumulation ratio for AUC24 was approximately 1.24.9,15 Distribution Absorption After oral administration of E2V, it is cleaved to 17␤-estradiol and valeric acid during absorption by the 40 According to the manufacturer, estradiol is 38% bound to sex hormone– binding globulin (SHBG) and 60% bound to albumin; the remaining 2% to 3% cir- Volume 34 Number 1 L.M. Borgelt and C.W. Martell culates in free form.9 No statistically significant changes in SHBG serum concentrations were observed during 28 days of medication administration (57.9 nmol/L [%CV, 42.9%] on day 1 and 81.5 nmol/L [%CV, 38.1%] on day 29).15 Serum cortisol-binding globulin (CBG) concentrations also mostly remained unchanged throughout treatment (45.5 ␮g/mL [%CV, 55.4%] on day 1 and 49.2 ␮g/mL [%CV, 24.8%] on day 29).15 Intravenous administration resulted in an apparent volume of distribution of approximately 1.2 L/kg.9 The product information indicates approximately 10% of circulating DNG is present in free form, with 90% found nonspecifically to albumin.9 Dienogest does not bind to SHBG or CBG. The volume of distribution at steady state of DNG is 46 L after 85 ␮g of 3 H-DNG is administered intravenously.9 Metabolism The product information states that estradiol undergoes extensive first-pass metabolism through the cytochrome P450 (CYP) 3A system in the liver, resulting in production of its metabolites estrone, estrone sulfate, or estrone glucuronide.9 A considerable amount of estradiol metabolism also occurs in the gastrointestinal mucosa. In total, ⬃95% of the oral dose becomes metabolized before entering the systemic circulation.9 Dienogest is extensively metabolized in the CYP3A4 system using hydroxylation and conjugation to form mostly inactive metabolites.9 Elimination The manufacturer reports that estradiol and its metabolites are predominantly excreted in the urine.9 Minimal (⬃10%) excretion occurs in the feces. The terminal half-life of estradiol is approximately 14 hours.9 Dienogest metabolites are primarily excreted renally; unchanged DNG is the dominating fraction in plasma.9 The terminal half-life of DNG is approximately 11 hours.9 pausal women. Studies evaluating the pharmacokinetics of E2V/DNG in patients with renal impairment were not found with this literature review, but renal dose adjustment is not likely to be necessary. Likewise, studies evaluating the pharmacokinetics in patients with liver impairment were not identified in this review; however, because the metabolism of steroid hormones may be affected with impaired liver function, the product information indicates that acute or chronic impairment of liver function may require discontinuation of COC use until liver function returns to normal.9 Obesity This literature review did not find that the safety and efficacy of E2V/DNG had been studied in women with a BMI ⬎30 kg/m2. Pregnancy and Lactation Estradiol valerate/dienogest is classified as a pregnancy category X agent, indicating that it is contraindicated in pregnancy. Women can rest assured that extensive epidemiologic studies have not demonstrated an increased risk of birth defects when COCs have been used before pregnancy or teratogenic effects when unintentionally taken during early pregnancy.3,9 If pregnancy is confirmed, COC use should be discontinued. Breastfeeding mothers should use other forms of contraception until the child has been weaned. Oral contraceptives containing estrogen can reduce milk production. In addition, small amounts of steroids and/or metabolites from the COC will present in breast milk. Drug-Drug Interactions Drug interactions with E2V/DNG may result in breakthrough bleeding and/or contraceptive failure. Although no relevant studies were identified in the literature search, the prescribing information indicates that the following interactions have been either reported with other COCs or studied in clinical trials with E2V/DNG.9 Special Populations Renal and Hepatic Impairment CYP3A4 Inducers The safety and efficacy of E2V/DNG have been established in women of reproductive age (18 –35 years). The manufacturer states that efficacy is expected to be similar for postpubertal adolescents ⬍18 years, but use of this medication before menarche is not recommended.9 This 4-phase formulation of E2V/DNG has not been studied and is not indicated in postmeno- Medications that induce CYP3A4 may decrease the effectiveness of COCs or increase breakthrough bleeding. Dienogest is a substrate of CYP3A4. Women who take medications that are strong CYP3A4 inducers should choose another form of contraception while using these inducers and for at least 28 days after discontinued use. Examples of medications that may result in de- January 2012 41 Clinical Therapeutics creased contraceptive efficacy include barbiturates, bosentan, carbamazepine, felbamate, griseofulvin, oxcarbazepine, phenytoin, rifampicin, St. John’s wort, and topiramate. The product information reports that coadministration of rifampicin with E2V/DNG tablets in postmenopausal women resulted in a 25% and 44% decrease in Cmax and AUC0 –24, respectively, for estradiol and a 52% and 83% decrease in the mean Cmax and AUC0 –24, respectively, for DNG.9 CYP3A4 Inhibitors According to the manufacturer, strong CYP3A4 inhibitors, such as ketoconazole, and moderate CYP3A4 inhibitors, such as erythromycin, have been found to increase hormone serum concentrations. Coadministration of E2V/DNG and ketoconazole resulted in a 65% and 57% increase of Cmax and AUC0 –24 for estradiol and a 94% and 186% increase of Cmax and AUC0 –24 for DNG at steady state.9 Coadministration of E2V/DNG and erythromycin resulted in a 51% and 33% increase of Cmax and AUC0 –24 for estradiol and a 33% and 62% increase of Cmax and AUC0 –24 for DNG.9 Other Agents Coadministration of HIV protease inhibitors has caused significant changes (increases or decreases) in the plasma levels of estrogen and progestin.9 Case reports of pregnancy have occurred with coadministration of antibiotics and COCs; however, clinical pharmacokinetic studies have not demonstrated consistent effects on plasma concentrations of steroids.21–27 The manufacturer reports that in vitro studies have not shown an inhibitory potential of DNG with human CYP enzymes at clinically relevant concentrations.9 Clinical Efficacy Ovulation Inhibition The initial efficacy of E2V/DNG was investigated in 2 prospective, randomized, open-label, Phase II dosefinding studies.19 Within these 2 studies, 4 variations of a 4-phasic E2V/DNG preparation were evaluated for their ability to inhibit ovulation. The studies were conducted sequentially. Study 1 evaluated the optimal treatment length of E2V/DNG. Study 2 aimed to determine the DNG dose necessary for ovulation inhibition. Conducted at 2 European centers (Germany and the Netherlands), the investigators enrolled healthy women aged 18 to 35 years and smokers ⱕ30 years. Exclusion criteria included contraindications for COC use, coexisting disease (metabolic and endocrine dis- 42 eases, unclassified genital bleeding, history of liver or vascular diseases), concurrent use of medications that could influence the study objectives (sex steroids), and drugs known to induce or inhibit liver enzymes. To undergo randomization, eligible women were required to demonstrate ovulation or a follicular diameter of ⱖ15 mm (assessed by transvaginal ultrasonography) on or before day 23 of a pretreatment observation cycle. Participants received study medication for up to 3 consecutive menstrual cycles of 28 days, without a pillfree interval. Use of the study medication was initiated on day 1 of menstrual bleeding during the first study cycle after a pretreatment observation cycle. Randomized participants were required to use nonhormonal contraceptive methods to protect against pregnancy throughout the study. The primary efficacy end point of ovulation inhibition in these 2 studies was assessed by the Hoogland and Skouby scoring system.28 In this 6-point scoring system, a Hoogland score of 5 is defined as a luteinized unruptured follicle, and a score of 6 (ovulation) is defined as a ruptured follicle-like structure of ⬎13 mm with a serum progesterone level of ⬎5 nmol/L and an E2 level of ⬎0.2 nmol/L. In both studies the primary efficacy end point was defined as the proportion of women with a Hoogland score of 5 or 6 at study cycle 2.19 The studies also assessed the proportion of women with a Hoogland score of 5 or 6 at cycle 3 and the proportion of women with resumption of ovulation during a posttreatment cycle. Study 1 randomized 196 women to either regimen 1A or regimen 2A (Table I). Women who took at least 1 study tablet and for whom at least 1 observation after dosing was available were included in the full analysis set (FAS). Ninety-six women in each regimen were included in the FAS. The mean (SD) age was 25.4 (4.1) years and 26.0 (4.4) years, mean (SD) BMI was 22.6 (3.1) and 22.8 (2.7) kg/m2, percentage of nullipara women was 80.2% and 75.0%, prevalence of current smoking was 35.4% and 36.5%, and OC use before study start was 44.8% and 40.6% for regimens 1A and 2A, respectively. The primary efficacy end point was measured at the end of cycle 2, and only women with a nonpersisting follicle-like structure ⬎13 mm (n ⫽ 31) were administered a third treatment cycle.19 Study 2 used the results from study 1 and adopted the treatment length of regimen 2A. The E2V dose was unchanged, but 2 different doses of DNG (both higher Volume 34 Number 1 L.M. Borgelt and C.W. Martell Table I. Results from 2 prospective trials evaluating ovulation inhibition.19* Regimen 1A: days 1–3: 3 mg E2V; days 4–7: 2 mg E2V/1 mg DNG; days 8-23: 2 mg E2V/2 mg DNG; days 24–25: 1 mg E2V; days 26-28: placebo 2A: days 1–2: 3 mg E2V; days 3–7: 2 mg E2V/1 mg DNG; days 8–24: 2 mg E2V/2 mg DNG; days 25–26: 1 mg E2V; days 27–28: placebo 2B: days 1–2: 3 mg E2V; days 3–7: 2 mg E2V/2 mg DNG; days 8–24: 2 mg E2V/3 mg DNG; days 25–26: 1 mg E2V; days 27–28: placebo 2C: days 1–2: 3 mg E2V; days 3–7: 2 mg E2V/3 mg DNG; days 8–24: 2 mg E2V/4 mg DNG; days 25–26: 1 mg E2V; days 27–28: placebo No. of Women Evaluated at Cycle 2 No. (%) of Women With Hoogland Score of 5 or 6 at Cycle 2 [90% CI, %] 92 10 (10.9) [5.53–16.21] 94 6 (6.4) [2.24–10.53] 96 3 (3.1) [0.20–6.05] 97 1 (1.0) [0.00–2.72] DNG ⫽ dienogest; E2V ⫽ estradiol valerate. *P values not reported. than that used in regimen 2A) were investigated. Study 2 randomized 210 women to either regimen 2B or 2C, and 100 and 103 women were included in the FAS, respectively. The mean (SD) age was 25.6 (3.7) years and 26.0 (4.2) years, mean (SD) BMI was 22.3 (2.9) and 22.2 (2.8) kg/m2, percentage of nullipara women was 89.0% and 79.6%, prevalence of current smoking was 33.0% and 34.0%, and OC use before study start was 44.0% and 48.5% for regimens 2B and 2C, respectively. In contrast to study 1, all participants in study 2 received 3 treatment cycles. The primary efficacy data are summarized in Table I. In study 1, 10 women (10.87%; 90% CI, 5.53%– 16.21%) in regimen 1A (n ⫽ 92) and 6 women (6.38%; 90% CI, 2.24%–10.53%) in regimen 2A (n ⫽ 94) had a Hoogland score of 5 or 6 in cycle 2. As stated earlier, based on the results of study 1, regimen 2A was selected as the template regimen for study 2. In study 2, 3 women (3.13%; 90% CI, 0.20%– 6.05%) in regimen 2B (n ⫽ 96) and 1 woman (1.03%; 90% CI, 0.00%– 2.72%) in regimen 2C (n ⫽ 97) had a Hoogland score of 5 or 6 in cycle 2. P values were not reported for these study results. In study 1, 8 of 15 women (53.3%) in regimen 1A and 5 of 16 women (31.3%) in regimen 2A had ovulations in treatment cycle 3. In study 2, 2 of 91 women January 2012 (2.2%) in regimen 2B and 1 of 95 women (1.05%) in regimen 2C ovulated in treatment cycle 3. Thirty-five to 44 women in each regimen were assessed during a posttreatment cycle, and return of ovulation occurred in 81.8% to 97.5% of these women. On the basis of the results of these 2 sequential studies, several regimens of E2V/DNG demonstrated efficient ovulation inhibition. The investigators’ targeted an ovulation rate of ⬍5% with an upper 90% CI limit of ⬍10%, based on research by Spona et al.29 Both regimens 2B and 2C achieved this efficacy target, and regimen 2B was found to be the lowest-effective dose/regimen for ovulation inhibition. Therefore, regimen 2B was used in subsequent Phase III investigations and, ultimately, in the currently available 4-phasic E2V/DNG COC product. Pregnancy Prevention A large, Phase III trial with a longer study period (20 cycles) was conducted by Palacios et al30 to evaluate the efficacy of 4-phasic E2V/DNG on prevention of pregnancy. The study was a multicenter, open-label, noncomparative study conducted in 50 centers across Europe (18 in Austria, 27 in Germany, 5 in Spain) for 2 years. Healthy women aged 18 to 50 years requesting contraception were eligible for study inclusion. Exclu- 43 Clinical Therapeutics Table II. Clinical efficacy of estradiol valerate/dienogest in a large Phase III trial.30* Efficacy End Point Pregnancies (during study treatment), no. Unadjusted Pearl Index (all pregnancies) (upper 95% CI limit) Adjusted Pearl Index (method failure pregnancies) (upper 95% CI limit) Cumulative failure rate at 20 cycles (Kaplan-Meier estimate) (95% CI) All Women (n ⫽ 1377) Women Aged 18-35 Years (n ⫽ 998) 13 12 0.73 (1.24) 0.94 (1.65) 0.34 (0.73) 0.40 (0.92) 0.0109 (0.0063–0.0188) 0.0142 (0.0080–0.0251) *P values not reported. Treatment was as follows days 1 and 2: 3 mg estradiol valerate (E2V); days 3 through 7: 2 mg E2V/2 mg dienogest (DNG); days 8 through 24: 2 mg E2V/3 mg DNG; days 25 and 26: 1 mg E2V; and days 27 and 28: placebo. sion criteria were generally consistent with the usual contraindications for OC use. Women were also excluded if they had a BMI ⬎30 kg/m2 and were smokers aged ⬎30 years. The study participants received a 28-day treatment regimen consisting of the following E2V/DNG doses: days 1 and 2: E2V 3 mg; days 3 through 7: E2V 2 mg/DNG 2 mg; days 8 through 24: E2V 2 mg/DNG 3 mg; days 25 and 26: E2V 1 mg; and days 27 and 28, placebo. The first dose was taken on the first day of withdrawal bleeding of the first treatment cycle, and the daily medication was continued for 20 cycles with no tablet-free interval between cycles. The primary efficacy outcome was the number of observed pregnancies, defined as unintended pregnancies during the treatment period. Contraceptive efficacy was estimated by calculating the Pearl Index (number of pregnancies per 100 women-years of exposure) and the corresponding upper limit of the 95% CI. Kaplan-Meier estimation was also used to calculate the cumulative failure rate on the basis of known pregnancies under treatment. Pregnancies were evaluated by an investigator for user or method failure. User failures were classified as pregnancies that resulted from 1 of the following scenarios: incorrect medication administration by the user, the user had taken a substance or experienced an concomitant illness that may have affected the oral contraception absorption, or the user experienced vomiting or diarrhea within 4 hours after tablet intake without intake of a second tablet. The information regarding tablet intake and other sources of suspected E2V/DNG failure were gathered from 44 participant diaries completed daily and/or additional documentation. The sample size was calculated based on the difference between the upper limit of the CI and point estimate not exceeding 1 with a probability of at least 90% when the Pearl Index was 1. This would obtain a 2-sided 95% CI for the Pearl Index. Data from 12,337 cycles of women aged 18 to 35 years were required. The number of women needed, assuming a dropout rate of 30%, was 1200 women aged 18 to 50 years (including 881 women aged 18 –35 years) to allow sufficient power. The study recruited 1391 women, and participants who received at least 1 dose of study medication and for whom at least 1 observation after admission was available were included in the FAS (n ⫽ 1377). The primary efficacy outcomes (Pearl Indices and KaplanMeier estimates) were assessed for the FAS and for the subgroup of women aged 18 to 35 years (n ⫽ 998). Overall, the mean (SD) age was 30.3 (7.9) years (26.2 [4.7] years for the women aged 18 –35 years), BMI was 22.8 (2.9) kg/m2, percentage sexually active was 99.5%, and prevalence of concurrent smoking was 19.8% (27.2% in the women aged 18 –35 years). A total of 13 pregnancies occurred during the study treatment, and 12 of these pregnancies occurred in the women aged 18 to 35 years (Table II). The 13 pregnancies occurred during an exposure time of 23,368 cycles, for an unadjusted Pearl Index of 0.73 (upper limit of the 95% CI, 1.24). Six of the 13 pregnancies were classified as method failure, for an adjusted Pearl Index of 0.34 (upper limit of the 95% CI, 0.73). In women Volume 34 Number 1 L.M. Borgelt and C.W. Martell aged 18 to 35 years, the 12 pregnancies occurred during an exposure of 16,608 cycles, for an unadjusted Pearl Index of 0.94 (upper limit of the 95% CI, 1.65). Five of the 12 pregnancies were classified as method failure, for an adjusted Pearl Index of 0.40 (upper limit of the 95% CI, 0.92). The Kaplan-Meier estimate for the cumulative failure rate during an exposure time of 20 cycles was 0.0109 (95% CI, 0.0063– 0.0188) in all study participants and 0.0142 (95% CI, 0.0080 – 0.0251) in the women aged 18 to 35 years. Overall, the Pearl Indices reported in this trial are similar to those reported in trials for EE-containing COCs.31–33 The efficacy of this formulation is comparable with reports of efficacy in trials involving estradiol-containing COCs.34 –37 A large, multicenter, randomized, double-blind European study with the primary outcome of comparing bleeding patterns (described in detail below) also reported efficacy data.38 This study evaluated the number of unintended pregnancies in women treated with 4-phasic E2V/DNG compared with monophasic EE/ LNG (20 ␮g/100 ␮g). During the 7 cycles of treatment, zero unintended pregnancies occurred in the group receiving E2V/DNG (FAS, n ⫽ 399), and 1 unintended pregnancy occurred in the group receiving EE/LNG (FAS, n ⫽ 399). No Pearl Index was reported. A pooled analysis of 3 large studies was conducted by Nelson et al39 in healthy women aged 18 to 50 years with a BMI ⬍30 kg/m2, and results were reported at the 2009 Annual Clinical Meeting of the American Congress of Obstetricians and Gynecologists. These trials were pooled and analyzed from North America and Europe, including 2 trials by Palacios et al30 and Ahrendt et al38 and 1 unpublished trial. The initial intent of the investigators was to perform their study during 13 cycles, but it was extended to a maximum of 28 cycles to gain additional safety data. The primary outcome was the number of observed pregnancies during treatment. A total of 2266 women received E2V/ DNG in the 3 trials. In the North American study, 7 pregnancies occurred, 6 of which were considered to have occurred “during treatment.” In the European open-label study, 14 pregnancies occurred, 11 of which were considered to have occurred “during treatment.” In the European comparative study, no pregnancies occurred. When looking at the pregnancies that were attributed to user failure (eg, nonuse or incorrect use) in women aged 18 to 35 years, the adjusted first-year Pearl Index was 0.72 (upper limit of 95% CI, January 2012 1.37). The unadjusted Pearl Index was 1.27 (upper limit of 95% CI, 2.06). In summary, currently available EE-containing COCs typically demonstrate a Pearl Index between 0 and 1, and the Pearl Index estimates in the identified studies of 4-phasic E2V/DNG ranged from 0.73 to 1.27 (unadjusted) to 0.34 to 0.72 (adjusted for method failure only).32,33,37,40,41 These comparable Pearl Indices, along with absolute rates of pregnancy data, indicated that the contraceptive efficacy of E2V/DNG is similar to other products within the COC medication class. Tolerability and Safety Profile Adverse Events In the 2 Phase II trials by Endrikat et al,19 a total of 12 women had treatment-related or “possibly related to treatment” adverse events that resulted in therapy discontinuation (2 in regimen 1A [ovarian cyst, breast pain], 2 in regimen 2A [edema, diarrhea], 5 in regimen 2B [depression, headache, worsening acne, eye irritation], and 3 in regimen 2C [emotional lability, headache]). Most treatment-related adverse events were typical of those occurring during COC use. The most frequently reported adverse events for the 4 different E2V/DNG regimens were headache, abdominal pain, acne, breast pain, dysmenorrhea, emotional lability, and nausea (refer to Table III for percentages of each regimen). The study by Palacios et al30 evaluated safety and tolerability via several methods. At the final study examination, patients were asked for subjective assessments, including overall satisfaction, future contraceptive choice, and emotional and physical well-being. The investigators collected adverse event reports throughout the 20 cycles for each patient and conducted physical and gynecologic examinations (including cervical smears and endometrial biopsies). Participants recorded bleeding with daily diaries and events and were compared to a 90-day reference period (data not reported).38 Palacios et al30 reported that 1074 of 1377 patients (78.0%) completed the 20-cycle study course. Treatment was discontinued in 140 participants (10.2%) because of adverse events. Metrorrhagia or breakthrough bleeding (1.7%), acne (1.0%), and weight increase (0.9%) were the most common reasons for treatment discontinuation. A total of 917 women (66.6%) reported at least 1 adverse event during the 45 Clinical Therapeutics 46 Table III. Adverse events reported in various clinical trials. Treatment-Related Adverse Events, No. (%) Study; Arm Abdominal Headache Pain Endrikat et al19; regimen 2B 18 (18.0) (n ⫽ 100) Palacios et al30; 4-phasic 26 (1.9) E2V/DNG (n ⫽ 1377) Ahrendt et al38; 4-phasic E2V/DNG (n ⫽ 399)‡ Ahrendt et al38; EE/LNG (n ⫽ 399)‡ Nelson et al39; 4-phasic E2V/DNG (n ⫽ 2266) Acne Breast Pain and/or Discomfort Dysmenorrhea Emotional Weight Lability Nausea Increase 12 (12.0) 9 (9.0) 6 (6.0) 10 (10.0) 7 (7.0) 7 (7.0) NR 36 (2.6) 67 (4.8) NR † NR 5 (1.3) 13 (3.3) 2 (0.5) NR NR 9 (2.3) 4 (1.0) 2 (0.5) NR NR 64 (2.8) 112 (4.9) 38 (1.7) NR NR Headache, NR 7 (1.8); migraine, 2 (0.5) Headache, NR 7 (1.8); migraine, 5 (1.3) 71 (3.1) 39 (1.7) NR Other NR Serious Events (n)* None 21 (1.5) Metrorrhagia, Presumed ocular histoplasmosis 26 (1.9) syndrome (1) Uterine leiomyoma (1) Focal nodular hyperplasia of the liver (1) Myocardial infarction (1) Deep vein thrombosis (1) 2 (0.5) Alopecia, Ruptured ovarian cyst (1) 3 (0.8) Autonomic nervous system imbalance (1) 4 (1.0) Alopecia, 4 (1.0) 34 (1.5) Metrorrhagia, 110 (4.9) Volume 34 Number 1 DNG ⫽ dienogest; E2V ⫽ estradiol valerate; NR⫽ not reported. *Only serious adverse events deemed by investigators as at least “possibly” treatment-related are reported here. † See Figure 5. ‡ For this trial, only the percentage of events categorized as “possibly, probably, or definitely” related to treatment are reported here. Breast cancer (1) L.M. Borgelt and C.W. Martell study treatment, and 272 women (19.8%) reported adverse events that the investigators considered “possibly, probably, or definitely related to treatment.” The most frequently reported treatment-related adverse events were breast pain (3.6%), acne (2.6%), headache (1.9%), metrorrhagia (1.9%), weight increase (1.5%), and breast discomfort (1.2%). Most adverse events were mild (Table III), and 59 “serious” adverse events were reported in 43 women. The investigators deemed 5 serious adverse events to be at least possibly related to study treatment. In the study by Ahrendt et al,38 which compared 4-phasic E2V/DNG with EE/LNG, patients were not asked to report adverse effects via direct inquisition by an investigator at each study visit. Women were given the opportunity to report adverse events at each visit, but concerns were to be generated from the patient without prompting or direct query from an investigator. During the 7-cycle study, 338 adverse events were reported in the study population (n ⫽ 399 in each treatment group, FAS), with 176 events (occurring in 108 women, 27.1%) in the E2V/DNG group and 162 events (occurring in 102 women, 25.6%) in the EE/ LNG group. Breast pain (3.8%), headache (2.5%), and vaginal infection (2.5%) were the most frequent adverse events in women treated with E2V/DNG, and acne (3.3%), headache (3.3%), and nasopharyngitis (1.8%) were the most frequent adverse events in women treated with EE/LNG. Adverse events considered possibly related to treatment occurred in 10.0% and 8.5% of women in the E2V/DNG and EE/LNG groups, respectively. P values were not reported for adverse events. The rate of discontinuation attributable to adverse events in either group was 3.3%, and bleeding events were not a cause of discontinuation of drug use.38 The events categorized as “possibly, probably, or definitely” related to treatment are presented in Table III. Eight serious adverse events were reported, 3 of which were considered to be possibly related to treatment (ruptured ovarian cyst and autonomic nervous system imbalance occurring in the same patient and breast cancer diagnosed in a 30-year-old nonsmoker 3 months after study termination). The pooled results of the 3 studies of E2V/DNG (with a total of 2266 patients) conducted by Nelson et al39 indicated that the most commonly reported adverse events related to E2V/DNG treatment were breast discomfort (4.9%), metrorrhagia (4.9%), and January 2012 headache (3.1%). The other adverse events that were reported in ⬎1% of patients were acne (2.8%), increased weight (1.5%), amenorrhea (1.7%), dysmenorrhea (1.7%), and abdominal pain (1.7%). No P values were reported in this analysis for these events. The adverse events associated with E2V/DNG in the studies described appear to be comparable to other available COCs.3,32,33,41 The most commonly reported treatment-related adverse events are metrorrhagia, breast discomfort, acne, and headache. Overall, the rates and nature of adverse events reported with E2V/DNG thus far do not suggest increased risk or warrant additional concerns versus other COCs. Death Palacios et al30 reported 2 deaths during the treatment period, but both were unrelated to the E2V/DNG treatment (accident and intracranial aneurism). No deaths were reported in the Phase II dose-finding studies by Endrikat et al,19 in the comparison study of E2V/DNG and EE/LNG by Ahrendt et al,38 or in any other studies performed in this review. Cardiovascular Safety No studies have been conducted to date to specifically look at the effects of E2V/DNG on cardiovascular safety. Two of the serious adverse events in the study by Palacios et al30 may have been complicated by extenuating circumstances. One of the 2 serious events presented as a leg deep vein thrombosis (DVT) in a 40-year-old woman after a sprained ankle. The DVT occurred 9 days after she completed treatment with E2V/DNG, and she had initiated contraception with depot medroxyprogesterone acetate in the meantime. The second adverse event was a myocardial infarction that occurred in a 47-year-old woman in violation of the study protocol (smoker aged 30 years at study entry). Endometrial Safety Bitzer et al42 investigated the endometrial safety of the 4-phasic E2V/DNG in healthy women aged 18 to 50 years. Endometrial biopsy specimens were taken at baseline (n ⫽ 283) and at cycle 20 (n ⫽ 218) in this multicenter, open-label, noncomparative trial. After 20 cycles of treatment, E2V/DNG showed endometrial safety by exhibiting no endometrial hyperplasia or malignancy, and 80.9% of women had an inactive (39.3%), atrophic (24.7%), or secretory (16.9%) endometrium (11.4% were considered “proliferative”) 47 Clinical Therapeutics with no harmful effects shown on endometrial histologic analysis. Cycle Control and Bleeding Some of the unacceptable adverse effects seen with estradiol-containing OCs have been poor cycle control and unacceptable menstrual bleeding.34,37,43,44 Such outcomes are important because unscheduled bleeding or spotting occurs in up to 30% to 50% of women starting use of COCs.3 To investigate these effects with the combination of E2V/DNG, a multicenter trial in 34 centers in Germany, the Czech Republic, and France was conducted.38 The study was a randomized, double-blind, double-dummy trial comparing a 4-phasic E2V/DNG regimen to a monophasic EE/LNG regimen. Trial participants were healthy women 18 to 50 years old seeking contraception. Women aged ⬎30 years could not smoke, and women aged 18 to 30 years were permitted to smoke up to 10 cigarettes daily. The exclusion criteria were consistent with the contraindications, special warnings, and precautions for OCs, including pregnancy or lactation, occurrence of ⬍3 menstrual cycles after childbirth, abortion, current use of an intrauterine device, obesity (BMI ⬎30 kg/m2), use of long-acting progestins within 6 months before study entry, hypersensitivity to any study drug ingredients, and known or suspected malignant or premalignant disease. Use of sex steroids was prohibited, but women switching from another OC were able to continue use until the end of the current cycle pack. Notably, normal menstrual cycles were not a study requirement. Study participants received either E2V/DNG (days 1–2: E2V, 3 mg; days 3–7: E2V, 2 mg/DNG, 2 mg; days 8 –24: E2V, 2 mg/DNG, 3 mg; days 2–26: E2V, 1 mg; days 27-28: placebo) or EE/LNG (days 1–21: EE, 20 ␮g/LNG, 100 ␮g; days 22–28: placebo) for 7 cycles of 28 days. Women began the daily treatment medication on the first day of menstrual bleeding (women not switching from another OC) or on the first day of withdrawal bleeding (women switching from another OC). Participants were required to complete daily diary cards to record tablet intake and bleeding events throughout the study. The primary outcomes were bleeding patterns and cycle control parameters. Measures of bleeding patterns included the number of bleeding or spotting days and the number and length of bleeding or spotting episodes; episodes were defined as bleeding or spotting 48 days with ⱖ2 days of no bleeding or spotting before or after the episode. Bleeding intensity was classified on the daily diary cards by the following scale: none, 1; spotting, 2; light, 3; normal, 4; or heavy, 5. For cycle control parameters, scheduled bleeding was defined as a bleeding or spotting episode that began during the hormone-free period or not more than 4 days before the progestin withdrawal. Absence of scheduled bleeding was defined as lack of bleeding until day 20 (E2V/ DNG) or day 17 (EE/LNG) of the following cycle. All other bleeding episodes were classified as unscheduled (intracyclic). In addition to these primary end points, unintended pregnancies and adverse events were recorded and evaluated. A total of 804 women were equally randomized to E2V/DNG or EE/LNG, and women who actually received treatment medication comprised the FAS (E2V/ DNG, n ⫽ 399; EE/LNG, n ⫽ 399). The mean (SD) age was 33.0 (9.0) years and 33.4 (8.8) years, mean (SD) BMI was 23.0 (2.9) and 23.2 (3.1) kg/m2, 99.5% and 100% were white, prevalence of OC use at screening was 91.7% and 91.0%, and prevalence of current smoking was 14.3% and 12.3% for the E2V/DNG and EE/LNG groups, respectively. The baseline incidence of dysmenorrhea in the preceding 6 months was 9.5% in the E2V/DNG group and 6.8% in the EE/LNG group. Statistical comparisons for the baseline characteristics were not reported. The study suggested effective cycle control with the E2V/DNG regimen during the 7-cycle treatment.38 Scheduled withdrawal bleeding per cycle was reported in 77.7% to 83.2% of women in the E2V/DNG group and in 89.5% to 93.8% of women in the EE/LNG group (P ⬍ 0.001 in each cycle). In accordance with this finding, more women in the E2V/DNG group experienced an absence of a scheduled withdrawal bleeding compared with the EE/LNG group (P ⬍ 0.0001). Within a given cycle, a mean of 19.4% of women taking E2V/DNG (range per cycle, 16.8%–22.3%) experienced no withdrawal bleeding and 7.7% of women taking EE/LNG (range per cycle, 6.2%-10.5%) experienced the same lack of scheduled withdrawal bleeding (P ⬍ 0.0001 for between group comparisons in all 7 cycles). The proportion of women who experienced an absence of scheduled withdrawal bleeding at least once during the 7 cycles was 56.9% in the E2V/DNG group and 37.8% in the EE/LNG group (no P value reported). Volume 34 Number 1 L.M. Borgelt and C.W. Martell Figure 4. Maximum intensity of scheduled withdrawal bleeding during cycles 1 through 7 in women treated with estradiol valerate/dienogest (E2V/DNG) and ethinyl estradiol/levonorgestrel (EE/LNG).38 Reprinted with permission from Contraception. Volume 80, number 5. Ahrendt H-J, Makalova D, Parke S, Mellinger U, Mansour D. Bleeding pattern and cycle control with an estradiol-based oral contraceptive: a seven-cycle, randomized comparative trial of estradiol valerate/dienogest and ethinyl estradiol/levonogestrel. Pages 436 – 444. Copyright 2009, with permission from Elsevier. The proportion of women in each treatment group who experienced unscheduled intracyclic bleeding per cycle was reported to be similar between the treatment groups (P ⬎ 0.05 per cycle). Approximately 14% (range, 10.5%–18.6%) of women in the E2V/DNG group reported intracyclic bleeding, and 12% (range, 9.9%–17.1%) in the EE/LNG group reported the same outcome. The highest rate of unscheduled intracyclic bleeding appeared to occur in the first cycle for both groups (no P value reported). The reported incidence of dysmenorrhea appeared to improve from baseline (9.5% in the E2V/DNG group and 6.8% in the EE/LNG group) to 0.5% in both groups during the study, but statistical analysis was not reported. In the first reference period (treatment days 1–90), women receiving E2V/DNG reported significantly fewer bleeding or spotting days compared with those receiving EE/LNG (17.3 [10.4] days [median, 16.0 days] vs 21.5 [8.6] days [median, 21.0 days]; P ⬍ 0.0001). Likewise for reference period 2 (treatment days 91–180), women in the E2V/DNG group reported significantly fewer bleeding or spotting days compared with women in the EE/LNG group (13.4 [9.3] days [median, 12.0 days] vs 15.9 [7.1] days [median, 15.0 days]; P ⬍ 0.0001). Scheduled withdrawal bleeding was shorter and lighter for women treated with E2V/DNG versus EE/ LNG. The mean length of withdrawal bleeding (during the 7 cycles) was 4.1 to 4.7 days (median, 4.0 days) in the E2V/DNG group and 5.0 to 5.2 days (median, 5.0 January 2012 days) in the EE/LNG group (P ⬍ 0.05 per cycle). On the 0- to 5-point scale for withdrawal bleeding intensity (with 5 classified as heavy bleeding), the mean score in the E2V/DNG-treated women was 3.2 to 3.3 per cycle (median score, 3; light bleeding). The mean score in the EE/LNG-treated women was 3.6 per cycle (median score, 4; normal bleeding). It appeared that a larger proportion of women in the E2V/DNG group reported the maximum withdrawal bleeding intensity to be spotting or light bleeding compared with women in the EE/LNG group (P values not reported); conversely, more women treated with EE/LNG reported the maximum withdrawal bleeding intensity to be normal or heavy bleeding (P values not reported) (Figure 4). Regarding intensity of unscheduled intracyclic bleeding, most women in both treatment groups (⬃75% per cycle) reported either spotting or light bleeding per cycle. The total proportion of women reporting heavy bleeding during cycles 1 to 7 was appeared to be higher in women treated with EE/ LNG versus E2V/DNG (4.0% vs 2.4%; P value not reported). In another analysis, data regarding bleeding from the 3 efficacy studies evaluating E2V/DNG were pooled and analyzed during 13 treatment cycles.39 The mean duration of withdrawal bleeding was 4.0 to 4.6 days in cycles 1 through 13. Withdrawal bleeding was most frequently considered to be spotting or light. The proportion of women with absent amenorrhea was 19.0% to 24.0% per cycle. Breakthrough bleeding 49 Clinical Therapeutics tended to decrease over time (23.3% in cycle 2 to 14.5% in cycle 13; P values not reported), and when it occurred, it was most often considered spotting or light. Heavy Menstrual Bleeding Two studies evaluated E2V/DNG specifically for its role in treating heavy and/or prolonged menstrual bleeding (HPMB).45,46 The multicenter, double-blind, randomized, placebo-controlled trials (one conducted in Europe and Australia and the other in the United States and Canada) had identical enrollment criteria and studied the same intervention, but they had different primary outcome measures. Both studies observed women aged ⱖ18 years during a 90-day run-in phase to confirm a diagnosis of prolonged bleeding (ⱖ2 episodes each lasting ⱖ8 days), frequent bleeding (⬎5 episodes with ⱖ20 bleeding days overall), or heavy bleeding (ⱖ2 episodes of blood loss ⱖ80 mL). Women whose menstrual bleeding was due to a recognizable or organic pelvic condition were not included in the study. Women who fit the study criteria were randomized in a 2:1 distribution to E2V/DNG (estrogen stepdown and progestin step-up during 26 days followed by 2 hormone-free days) or placebo for 196 days. Data from the 90-day run-in phase were compared with data from the last 90 days of treatment for each treatment groups. The primary end point in the North American study was complete resolution of abnormal menstrual symptoms, and secondary end points were changes in menstrual blood loss (MBL) volume and iron metabolism.46 A total of 190 women were randomized to E2V/ DNG (n ⫽ 120) or placebo (n ⫽ 70), with 136 completing the study (E2V/DNG, n ⫽ 85; placebo, n ⫽ 51). Excluding women with insufficient menstrual bleeding data, significantly more women achieved the primary outcome in the E2V/DNG group (35/80; 43.8% [95% CI, 32.7%–55.3%]) compared with the placebo group (2/48; 4.2% [95% CI, 0.5%–14.3%]) (P ⬍ 0.001). This finding remained statistically significant when considering missing data as nonresponders (intent-to-treat analysis) with complete resolution of abnormal menstrual symptoms in 35 of the 120 women (29.2%) in the E2V/DNG group and 2 of the 170 women (2.9%) in the placebo group (P ⬍ .001). The reduction in MBL (from the initial 90-day run-in phase to the final 90-day efficacy phase) was greater in the E2V/DNG-treated women compared with those 50 given placebo (⫺353 vs ⫺130 mL; P ⬍ 0.001), and the change was evident beginning in cycle 2. Finally, hemoglobin, hematocrit, and ferritin levels improved in the E2V/DNG group but not in the placebo group. The mean change from baseline to study end in the E2V/ DNG versus placebo group was ⫹0.6 versus ⫹0.1 g/dL (P ⫽ 0.004), ⫹1.4% versus ⫺0.05% (P ⫽ 0.001), and ⫹2.9 versus ⫺0.4 ng/mL (P ⫽ 0.011), respectively. The primary end points in the European and Australian study were MBL and iron metabolism parameters, and a total of 231 women were randomized to E2V/DNG (n ⫽ 149) or placebo (n ⫽ 82).45 Similar to the North American study, women treated with E2V/ DNG experienced significantly less MBL. The adjusted mean difference in MBL with E2V/DNG versus placebo was ⫺373 mL (95% CI, ⫺490 to ⫺255 mL; P ⬍ 0.0001). The iron metabolism parameters (hemoglobin and ferritin concentrations) were significantly improved in the E2V/DNG group, and the adjusted mean difference in with E2V/DNG versus placebo were ⫹0.6 g/dL (95% CI, 0.3–1.0 g/dL; P ⬍ 0.0001) and ⫹8.2 ng/mL (95% CI, 3.5–12.9 ng/dL; P ⬍ 0.002), respectively. Work Productivity and Activity Impairment in HPMB Aside from the physiologic and medical problems caused by HPMB (eg, bleeding symptoms, MBL), excessive menstrual bleeding may negatively affect a woman’s quality of life. In women with HPMB, 1 study evaluated the effect of E2V/DNG on quality of life and daily functioning.47 The study was a post hoc analysis of the randomized, placebo-controlled trial of E2V/ DNG for treatment of HPMB conducted in North America.46 Women were randomized to E2V/DNG (estrogen step-down and progestin step-up for 26 days followed by 2 hormone-free days) or placebo for 196 days (7 menstrual cycles). The objective of this analysis was to estimate the effect of HPMB on work productivity and impairment of activities of daily living (ADL) and the improvement with E2V/DNG during the 7-cycle treatment. These outcomes were measured by a work productivity and activities impairment questionnaire that study patients completed at baseline and treatment conclusion. The questionnaire used a 10point Likert scale with increased impairment determined with higher values. The mean (SD) baseline work productivity impairment score was 4.1 (0.2) in the United States and 4.0 (0.5) in Canada. At the end of treatment, women Volume 34 Number 1 L.M. Borgelt and C.W. Martell treated with E2V/DNG appeared to have a lower impairment score and a greater reduction in impairment from baseline (United States: 2.2 [0.2], ⫺46.2%; Canada: 2.1 [0.3], ⫺47.3%) compared with the placebotreated women (United States: 3.6 [0.4], ⫺13.1%; Canada: 3.3 [0.5], ⫺16.1%). Statistical analysis was not reported. The mean (SD) baseline ADL impairment score was 5.1 (0.2) and 4.6 (0.5) in the United States and Canada, respectively. Similar to the work productivity result, the ADL impairment score at treatment end appeared lower and the reduction from baseline was greater in women treated with E2V/ DNG (United States: 2.4 [0.2], ⫺53.0%; Canada: 2.0 [0.2], ⫺56.2%) compared with the placebotreated women (United States: 3.8 [0.4], ⫺24.8%; Canada: 3.3 [0.5], ⫺28.0%) (P values not reported). Although the clinical significance of a 2- to 3-point reduction on the Likert scale may be unclear, E2V/ DNG did provide an approximately 50% reduction in patient-scored impairment. When the cycle control and bleeding data for the 4-phasic E2V/DNG are considered together, this COC regimen appears to have a desirable effect on these outcomes. Although previous formulations of estradiol-containing treatments produced bleeding that may have limited its use, E2V/DNG has shown reductions in menstrual bleeding severity, duration, and volume.34,37,43,44 Many women treated with this therapy experienced a lack of withdrawal bleeding after several mediation cycles. The regimen appears to reduce HPMB blood loss; stabilize hemoglobin, hematocrit, and ferritin levels; and reduce patient-scored HPMBassociated impairment of work productivity and ADL. In the trial that compared E2V/DNG and EE/LNG, E2V/DNG seemed to provide more favorable bleeding outcomes and overall cycle control. Effect on Various Metabolic and Hemostatic Parameters A double-blind, randomized, controlled, 4-arm, bicentric clinical study evaluated the effect of 4 OCs on lipid metabolism, hemostatic parameters, thyroid hormones, adrenal parameters, blood pressure markers, sex hormones, and serum-binding globulins.48 –51 Each group was composed of 25 patients with a mean (SD) age of 26.1 (4.5) years and a BMI of 21.9 (2.8), and 91 women completed the study. Women took 6 cycles of 1 of the following monophasic 21-day regimens: 30 ␮g of EE and 2 mg of DNG; 20 ␮g of EE and 2 mg of January 2012 DNG; 10 ␮g of EE, 2 mg of E2V, and 2 mg of DNG; or 20 ␮g of EE and 100 ␮g of LNG. Blood samples were taken on days 21 to 26 during a control cycle and then on days 18 to 21 during treatment cycles 1, 3, and 6. Regarding lipid metabolism, the DNG-containing regimens resulted in changes in lipid metabolism than those containing LNG.50 For example, compared with regimens containing DNG, LNG had lower high-density lipoprotein cholesterol 2 levels (P ⬍ 0.05) (without significant changes in highdensity lipoprotein cholesterol levels) and higher lowdensity lipoprotein cholesterol levels (P ⬍ 0.05) at cycle 6. Compared with baseline, triglyceride levels remained unchanged with the LNG regimen; however, triglycerides increased with the DNG regimens (P ⬍ 0.05). Compared with the control cycle, hemostatic parameters changed in cycle 6 for all 4 regimens with an increase in fibrinogen (P ⬍ 0.01), D-dimer (P ⬍ 0.01), plasminogen (P ⬍ 0.01), plasmin-antiplasmin complex (P ⬍ 0.01), and protein C activity (P ⬍ 0.01). Significant decreases in tissue plasminogen activator (P ⬍ 0.01) and plasminogen activator inhibitor (P ⬍ 0.01) in cycle 6 occurred versus the control cycle.48 Treatment with EE/LNG significantly increased the levels of free protein S compared with those women treated with DNG regimens (P ⬍ 0.01). The 3 DNGcontaining regimens increased factor VII activity at cycle 6 versus the control cycle (P ⬍ 0.01); no change occurred with the LNG-containing regimen. When evaluating thyroid, adrenal, and blood pressure parameters, all 4 regimens significantly increased total triiodothyronine (P ⬍ 0.01), thyroxine (P ⬍ 0.01), and cortisol (P ⬍ 0.01) at cycle 6 compared with baseline; however, free concentrations of these hormones were not or were only slightly affected.51 Angiotensin II decreased significantly for all 4 regimens in the sixth cycle (P ⬍ 0.01); aldosterone and endothelin 1 concentrations were not changed. Lastly, COC regimens containing DNG caused a higher increase in SHBG (P ⬍ 0.01) and thyroxine-binding globulin (TBG) (P ⬍ 0.01) compared with the regimen containing LNG.49 The lowest increase in CBG occurred in the regimen containing E2V, which suggests a lesser hepatic effect compared with EE (P ⬍ 0.01). All 4 regimens produced significant increases in SHBG (P ⬍ 0.01), CBG (P ⬍ 0.01), and TBG (P ⬍ 0.01), whereas decreases occurred in free testosterone (P ⬍ 0.01) and dehydroepiandrosterone sulfate (P ⬍ 0.01) compared with the control cycle.49 These findings are provided as a repre- 51 Clinical Therapeutics sentation of the components in E2V/DNG, but the study regimens did not contain the exact dosage in the currently available COC multiphasic product. Klipping et al52 reported similar results on hemostatic parameters with the 4-phasic regimen of E2V/ DNG. In this study, 34 healthy women aged 18 to 50 years were randomized to E2V/DNG (n ⫽ 16) or EE/ LNG (n ⫽ 16) in a crossover, open-label manner during 3 cycles. The primary outcomes were intraindividual absolute changes in prothrombin fragments 1 and 2 and D-dimer from baseline to cycle 3. A total of 29 women were part of the FAS because they received at least 1 dose of their assigned treatment. Treatment with E2V/DNG produced no intraindividual change in levels of prothrombin 1 and 2, and although a slight increase was seen with EE/LNG, the differences between the 2 treatments were not significant. A smaller increase in D-dimer occurred in the E2V/DNG group (from 203.0 [94.1] ng/mL to 237.4 [101.6] ng/mL) compared with the EE/LNG group (from 201 [73.5] ng/mL to 352.6 [217.8] ng/mL), which was statistically significant (P ⫽ .01; 95% CI, ⫺217.22 to ⫺11.77). Although these studies revealed similar or less pronounced effects with E2V/DNG, it is important to use caution when interpreting these results because these parameters have not been shown to predict the occurrence of thromboembolic events. Satisfaction The randomized, double-blind study by Ahrendt et al38 comparing 4-phasic E2V/DNG to monophasic EE/LNG measured patient satisfaction as a study outcome. Patients were randomized to 1 of the treatment regimens for 7 treatment cycles in this investigation of bleeding and cycle control. At the final examination at the study conclusion, patients were asked to report their satisfaction with the treatment that they had received for the previous 7 menstrual cycles. The assessment used the following categories: very satisfied, somewhat satisfied, neither satisfied or dissatisfied, dissatisfied, and very dissatisfied. A similar number of women in each treatment group reported being very satisfied or somewhat satisfied with treatment (79.4% in the E2V/DNG group and 79.9% in the EE/LNG group). Slightly more women in the E2V/DNG versus EE/LNG group responded that they were very satisfied with treatment (39.8% vs 35.3%) (no P values given). In the trial by Palacios et al,30 79.5% of women were very satisfied or satisfied (52.1% and 27.4%, re- 52 Figure 5. Physical and emotional well-being of women taking estradiol valerate/dienogest.30 Reprinted with permission from European Journal of Obstetrics & Gynecology and Reproductive Biology. Volume 149, number 1. Palacios S, Wildt L, Parke S, Machlitt A, Romer T, Bitzer J. Efficacy and safety of a novel oral contraceptive based on oestradiol (oestradiol valerate/ dienogest): A Phase III trial. Pages 57– 62. Copyright 2010, with permission from Elsevier. spectively) and 7.4% were dissatisfied or very dissatisfied (6.5% and 0.9%, respectively) with the 4-phasic E2V/DNG regimen (no P values reported). More than 86% of women also rated their physical and emotional well-being on study treatment as the same, better, or much better, compared with pretreatment status (Figure 5). Cost and Economic Considerations No cost-effectiveness analyses or other pharmacoeconomic evaluations of E2V/DNG were identified through the literature search. The approximate cost to patients in the United States is $84.99 for a pack of 28 tablets (26 active tablets and 2 inactive tablets).53 Although the clinical efficacy of COCs is comparable, several factors go into the varying costs. For example, availability of generic versions of COCs tend to significantly reduce cost. Formulary availability and coverage also affect the cost for individual patients. This COC is currently only available as a brand name product in the United States and is roughly equivalent than Volume 34 Number 1 L.M. Borgelt and C.W. Martell other branded contraceptives (eg, EE/drospirenone* is $92.00 for 28 tablets and EE/norgestimate† is $89.99 for 28 tablets). DISCUSSION Results from clinical trials demonstrate that E2V/DNG is an efficacious COC that is well tolerated and may in fact have a better bleeding profile that other COCs containing LNG. Two new hormone components are introduced with this COC formulation. Estradiol valerate is metabolized to natural 17␤-estradiol. This product appears to yield comparable results regarding metabolic (eg, cholesterol concentrations) and hemostatic (eg, fibrinogen, protein C activity) parameters when compared with COCs containing LNG; sex hormone parameters (eg, SHBG, TBG) show significant changes. It is reassuring that the data to this point show no increased thrombotic risk with the progestogen used. Comparative trials of E2V/DNG with EE/LNG, a criterion standard product for safety, have shown similar effects on the safety profile.17,30,39,54 However, these studies have not indicated an improved safety profile over existing COCs (eg, DVT). The current contraindications, precautions, and warnings that exist for other COC formulations also apply to this formulation. Future epidemiologic studies will help determine whether a COC containing E2V/DNG will lower these risks. The unique 4-phase formulation offers a potential benefit for women with HMB, a condition that can adversely affect women’s work, productivity, and quality of life. Although the formulation is not indicated for this purpose, a decreased number of bleeding days or absence of withdrawal bleeding is likely to occur with this formulation. The reduced bleeding profile may make this formulation ideal for women of older reproductive age (eg, perimenopausal) because they typically have a higher rate of bleeding irregularity with heavier or longer menstrual cycles.55 It may also be advantageous for this population because adherence is extremely important with E2V/DNG.56 The rules around missing pills are much more complicated than typical pills because of varying doses every few days. Clear instructions about how to handle missed doses are included in each package insert but will need *Trademark: Yaz® (Bayer Healthcare Pharmaceuticals Inc., Wayne, New Jersey). † Trademark: Ortho Tri-Cyclen Lo® (Janssen Pharmaceuticals, Raritan, New Jersey). January 2012 to be a point of education for users. This may not be an ideal pill for a first-time user for several reasons, including the complications around missed doses and the vast number of COCs that are effective, tolerable, and available generically (less expensive). Reducing the adverse effects of breakthrough bleeding and/or spotting may lead to better adherence. Most women will welcome less bleeding with a COC, although some prefer the comfort of a monthly menstrual cycle to ensure they are not pregnant. Women using this formulation may opt to use a pregnancy test when withdrawal bleeding does not occur to confirm that the COC is effective. Although this new drug review attempted to provide a thorough report of the data for this new COC product, this review has some limitations. Publication bias may be present as a result of the recent development and approval of this product, and published studies are limited. Our exclusion of non-English studies may contribute to this bias, but an attempt was made to include evidence from outside the United States. An evidence grading system was not used, but the studies with a higher level of evidence (ie, Phase III randomized controlled trials) were given more focus within the review. Finally, some non–peer- reviewed sources were incorporated into the review, including manufacturer data and meeting abstracts. The lack of published data for certain topics, particularly pharmacokinetics, necessitated inclusion of manufacturer data, and meeting abstracts were included to broaden the evidence pool. Conclusions based solely on product information or abstracts were minimized. CONCLUSIONS This new COC, E2V/DNG, is a unique formulation with 2 new hormonal components not previously available in the United States. Data to date indicate that it offers efficacy, tolerability, and a satisfactory safety profile. The estradiol in combination with DNG in a 4-phase regimen offers a favorable bleeding profile. The medication may not be considered an initial COC choice for first-time users because of cost and adherence issues but may be beneficial for women who have experienced bleeding adverse effects with other COCs and are looking for shorter and/or lighter menstrual cycles. Epidemiologic studies will be needed to determine whether other biomarkers result in clinically significant different outcomes among E2V/DNG and other available COCs. 53 Clinical Therapeutics ACKNOWLEDGMENTS Both authors contributed equally to the literature search, data interpretation, and writing of the manuscript. 16. 17. CONFLICTS OF INTEREST The authors have indicated that they have no conflicts of interest regarding the content of this article. 18. 19. REFERENCES 1. Dhont M. History of oral contraception. Eur J Contracept Reprod Health Care. 2010;15(Suppl 2):S12–S18. 2. Djerassi C. Chemical birth of the pill. Am J Obstet Gynecol. 2006;194:290 –298. 3. Nelson A. Combined oral contraceptives. In: Hatcher RA, Trussel J, Nelson AL, et al, eds. Contraceptive Technology. 19th ed. New York, NY: Ardent Media Inc; 2007:193–270. 4. Barbieri R. In the news, now on the shelf: a novel estradiol-based OC. OBG Management. 2010;22:6, 10 –11. www.obgmanagement.com. Accessed January 4, 2011. 5. Alsina JC. After 50 years of ethinylestradiol, another oestrogen in combined oral contraceptives. Eur J Contracept Reprod Health Care. 2010;15:1–3. 6. Hoffmann H, Moore C, Zimmermann H, et al. Approaches to the replacement of ethinylestradiol by natural 17betaestradiol in combined oral contraceptives. Exp Toxicol Pathol. 1998;50:458 – 464. 7. Hoffmann H, Moore C, Kovacs L, et al. Alternatives of the replacement of ethinylestradiol by natural 17beta-estradiol in dienogest-containing oral contraceptives. Drugs Today. 1999;35:105–113. 8. Sitruk-Ware R, Nath A. The use of newer progestins for contraception. Contraception. 2010;82:410 – 417. 9. Natazia (estradiol valerate/dienogest) [package insert]. Wayne, NJ. Bayer HealthCare Pharmaceuticals; May2010. 10. Mishell DR, Jr.Rationale for decreasing the number of days of the hormone-free interval with use of low-dose oral contraceptive formulations. Contraception. 2005;71:304 –305. 11. Read CM. New regimens with combined oral contraceptive pills—moving away from traditional 21/7 cycles. Eur J Contracept Reprod Health Care. 2010;15(Suppl 2):S32–S41. 12. Guttmacher Institute. In brief: facts on contraceptive use in the United States, 2010. http://www.guttmacher. org/pubs/fb_contr_use.html. Accessed January 4, 2011. 13. Speroff L, Fritz MA. Oral contraception. In: Clinical Gynecologic Endocrinology and Infertility. 7th ed. Philadelphia, Pa: Lippincott Williams & Wilkins; 2005:861–942. 14. Jensen JT. Evaluation of a new estradiol oral contraceptive: estradiol valerate and dienogest. Expert Opin Pharmacother. 2010;11:1147–1157. 15. Zeun S, Lu M, Uddin A, Zeiler B, Morrison D, Blode H. Pharmacokinetics of an oral contraceptive containing 54 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. oestradiol valerate and dienogest. Eur J Contracept Reprod Health Care. 2009;14:221–232. Perez-Campos EF. Ethinylestradiol/dienogest in oral contraception. Drugs. 2010;70:681– 689. Hoy SM, Scott LJ. Estradiol valerate/dienogest: in oral contraception. Drugs. 2009;69:1635–1646. Dusterberg B, Nishino Y. Pharmacokinetic and pharmacological features of oestradiol valerate. Maturitas. 1982;4:315–324. Endrikat J, Parke S, Trummer D, Schmidt W, Duijkers I, Klipping C. Ovulation inhibition with four variations of a four-phasic estradiol valerate/dienogest combined oral contraceptive: results of two prospective, randomized, open-label studies. Contraception. 2008;78:218 –225. Oettel M, Breitbarth H, Elger W. The pharmacological profile of dienogest. Eur J Contracept Reprod Health Care. 1999;4(Suppl 1):2–13. Dickinson BD, Altman RD, Nielsen NH, Sterling ML. Drug interactions between oral contraceptives and antibiotics. Obstet Gynecol. 2001;98:853–860. Sparrow MJ. Pill method failure. N Z Med J. 1987;100:102–105. Kakouris H, Kovacs GT. Pill failure and non-use of secondary precautions. Br J Fam Plan. 1992;18:41– 44. Kovacs GT, Riddoch G, Duncombe P, et al. Inadvertent pregnancies in oral contraceptive users. Med J Aust. 1989; 150:549 –551. Neely J, Abate M, Swinker M, D’angio R. The effect of doxycycline on serum levels of ethinyl estradiol, norethindrone, and endogenous progesterone. Obstet Gynecol. 1991;77: 416 – 420. Murphy AA, Zacur HA, Charache P, Burkman RT. The effect of tetracycline on levels of oral contraceptives. Am J Obstet Gynecol. 1991;164(1 Pt 1):28–33. Friedman CI, Huneke AL, Kim MH. The effect of ampicillin on oral contraceptive effectiveness. Obstet Gynecol. 1980;55:33–37. Hoogland HJ, Skouby SO. Ultrasound evaluation of ovarian activity under oral contraceptives. Contraception. 1993; 47:583–590. Spona J, Feichtinger W, Kindermann C, Wunsch C, Brill K. Inhibition of ovulation by an oral contraceptive containing 100 micrograms levonorgestrel in combination with 20 micrograms ethinylestradiol. Contraception. 1996;54:299–304. Palacios S, Wildt L, Parke S, Machlitt A, Romer T, Bitzer J. Efficacy and safety of a novel oral contraceptive based on oestradiol (oestradiol valerate/dienogest): a Phase III trial. Eur J Obstet Gynecol Reprod Biol. 2010;149:57– 62. The safety and contraceptive efficacy of a 24-day low-dose oral contraceptive regimen containing gestodene 60 microg and ethinylestradiol 15 microg. Eur J Contracept Reprod Health Care. 1999;4(Suppl 2):9 –15. Ahrendt HJ, Nisand I, Bastianelli C, et al. Efficacy, acceptability and tolerability of the combined contraceptive ring, NuvaRing, compared with an oral contraceptive contain- Volume 34 Number 1 L.M. Borgelt and C.W. Martell 33. 34. 35. 36. 37. 38. 39. 40. ing 30 microg of ethinyl estradiol and 3 mg of drospirenone. Contraception. 2006;74:451– 457. Bachmann G, Sulak PJ, SampsonLanders C, Benda N, Marr J. Efficacy and safety of a low-dose 24-day combined oral contraceptive containing 20 micrograms ethinylestradiol and 3 mg drospirenone. Contraception. 2004;70:191–198. World Health Organization Task Force on Oral Contraception. A randomized, double-blind study of two combined oral contraceptives containing the same progestogen, but different estrogens. Contraception. 1980;21:445– 459. Astedt B, Jeppsson S, Liedholm P, Rannevik G, Svanberg L. Clinical trial of a new oral contraceptive pill containing the natural oestrogen 17 beta-oestradiol. Br J Obstet Gynaecol. 1979;86:732–736. Hirvonen E, Allonen H, Anttila M, et al. Oral contraceptive containing natural estradiol for premenopausal women. Maturitas. 1995;21:27–32. Serup J, Bostofte E, Larsen S, Westergaard J, Lebech PE. Natural oestrogens for oral contraception. Lancet. 1979;2:471– 472. Ahrendt HJ, Makalova D, Parke S, Mellinger U, Mansour D. Bleeding pattern and cycle control with an estradiol-based oral contraceptive: a seven-cycle, randomized comparative trial of estradiol valerate/dienogest and ethinyl estradiol/levonorgestrel. Contraception. 2009;80:436 – 444. Nelson A, Sampson-Landers C, Parke S, Jensen JT. Efficacy of estradiol valerate/dienogest OC: results of 3 large studies in North America and Europe. Abstract and poster presented at: 57th Annual Clinical Meeting of the American Congress of Obstetricians and Gynecologists; May 2– 6, 2009; Chicago, Ill Anttila L, Backhmann G, Hernadi L, Kunz M, Marr J, Klipping C. Contraceptive efficacy of a combined oral contraceptive containing ethinylestradiol 20 mcg/drosperinone 3 mg ad- January 2012 41. 42. 43. 44. 45. 46. 47. ministered in a 24/4 regimen: a pooled analysis of four open-label studies. Eur J Obstet Gynecol Reprod Biol. 2011;155:180 –182. Dando TM, Curran MP. Low-dose ethinylestradiol/levonorgestrel. Drugs. 2005;65:2299 –2306. Bitzer J, Parke S, Roemer T, Serrani M. Endometrial safety of an oral contraceptive containing estradiol valerate and dienogest. Int J Womens Health. 2011;3:127–132. Wenzl R, Bennink HC, van Beek A, Spona J, Huber J. Ovulation inhibition with a combined oral contraceptive containing 1 mg micronized 17 beta-estradiol. Fertil Steril. 1993; 60:616 – 619. Schubert W, Cullberg G. Ovulation inhibition with 17 beta-estradiol cyclooctyl acetate and desogestrel. Acta Obstet Gynecol Scand. 1987;66:543–547. Fraser I, Zeun S, Machlitt A, Mellinger U. A novel oral contraceptive comprising estradiol valerate/dienogest for the treatment of heavy and/or prolonged menstrual bleeding without organic cause: a doubleblind, randomised, placebo-controlled trial: World Congress of the International Federation of Gynecology and Obstetrics (FIGO) 4 –9 October 2009. Int J Gynecol Obstet. 2009; 107(Suppl 2):S183. Jensen J, Parke S, Mellinger U, Machlitt A, Fraser I. Effective treatment of heavy menstrual bleeding with estradiol valerate and dienogest: a randomized controlled trial. Obstet Gynecol. 2011;117:777–787. Wasiak R, Filonenko A, Jensen J, Law A, Jeddi M, Stull D. Impact of estradiol valerate/dienogest (E2V/DNG) on work productivity (WP) and activities of daily living (ADL) impairment in North American women with heavy and/or prolonged menstrual bleeding (HPMP). Paper pre- 48. 49. 50. 51. 52. 53. 54. 55. 56. sented at: American Society for Reproductive Medicine (ASRM) 66th Annual Meeting; October 23– 27, 2010; Denver, Colo. Wiegratz I, Lee JH, Kutschera E, Winkler UH, Kuhl H. Effect of four oral contraceptives on hemostatic parameters. Contraception. 2004;70:97–106. Wiegratz I, Kutschera E, Lee JH, et al. Effect of four different oral contraceptives on various sex hormones and serum-binding globulins. Contraception. 2003;67:25–32. Wiegratz I, Lee JH, Kutschera E, et al. Effect of dienogest-containing oral contraceptives on lipid metabolism. Contraception. 2002;65:223–229. Wiegratz I, Kutschera E, Lee JH, et al. Effect of four oral contraceptives on thyroid hormones, adrenal and blood pressure parameters. Contraception. 2003;67:361–366. Klipping C, Duijkers I, Parke S, et al. Hemostatic effects of a novel estradiolbased oral contraceptive: an openlabel, randomized, crossover study of estradiol valerate/dienogest versus ethinylestradiol/levonorgestrel. Drugs R D. 2011;11:159 –170. Drug prices and information, 2011. http://www.drugstore.com. Accessed July 11, 2011. vanHylckamaVliegA,HelmerhorstFM, VandenbrouckeJP,DoggenCJ,Rosendaal FR. The venous thrombotic risk of oral contraceptives, effects of oestrogen dose and progestogen type: results of the MEGA case-control study. BMJ. 2009;339:b2921. Nelson A, Stewart F. Menopause and perimenopausal health. In: Hatcher RA, Trussel J, Nelson AL, et al, eds. Contraceptive Technology. New York, NY: Ardent Media Inc; 2007:699–745. Rosenberg M, Waugh MS. Causes and consequences of oral contraceptive noncompliance. Am J Obstet Gynecol. 1999;180(2 Pt 2):276 –279. Address correspondence to: Laura M. Borgelt, PharmD, Departments of Clinical Pharmacy and Family Medicine, University of Colorado, 12850 E. Montview Blvd., V20-2124. Aurora, CO 80045. E-mail: [email protected] 55