Westhoff, Carolyn MD, MSc; Kaunitz, Andrew M. MD; Korver, Tjeerd PhD; Sommer, Werner MSc; Bahamondes, Luis MD; Darney, Philip MD, MSc; Verhoeven, Carole PhD
Combined oral contraceptives are the most popular method of reversible hormonal contraception in the United States and Europe.1 Prior attempts to replace ethinyl estradiol (E2) with 17β-E2 and other estrogens failed owing to poor cycle control.2,3 Recently, the combination of nomegestrol acetate and 17β-E2 (identical to endogenous estrogen) was introduced in a new monophasic combined oral contraceptive4–6 approved by the European Medicines Agency in 2011. Nomegestrol acetate is a selective progestin structurally related to progesterone that exhibits strong antigonadotropic activity and moderate antiandrogenic properties and does not possess estrogenic, androgenic, glucocorticoid, or mineralocorticoid activity.7,8 Together, nomegestrol acetate and 17β-E2 provide effective inhibition of ovulation with acceptable cycle control and have minimal effects on markers of hemostasis and endocrine function.4,5,9,10
We report the results of a phase 3 trial conducted in the Americas that was run in parallel with a second trial of similar design and size6 in Asia, Australia, and Europe. In our study, we assessed the efficacy, cycle control, safety, and tolerability of nomegestrol acetate and 17β-E2 in healthy, fertile female volunteers in comparison with an available monophasic combined oral contraceptive containing drospirenone 3.0 mg and ethinyl E2 30 micrograms, the leading oral contraceptive in the United States at the time this trial was designed (2005).
MATERIALS AND METHODS
This randomized, open-label, multicenter trial (NCT00413062) evaluated contraceptive efficacy, cycle control, safety, and tolerability of nomegestrol acetate (2.5 mg) and 17β-E2 (1.5 mg) as well as evaluation of acne and body weight. These characteristics were evaluated in comparison with a monophasic combined oral contraceptive comprised of drospirenone (3.0 mg) and ethinyl E2 (30 micrograms). The study recruited volunteers from 89 gynecology and general practitioner practices in the United States, Canada, Argentina, Brazil, Chile, and Mexico. The study protocol was approved by the independent ethics committee of each study center and was conducted in compliance with the current standards and principles of the Declaration of Helsinki and the International Conference on Harmonization Guideline for Good Clinical Practice. All women provided written informed consent before participating. Each study center planned to recruit 20–60 women. Volunteers were reimbursed only for their travel costs and received the study medication at no cost.
Healthy, sexually active women age 18 to 50 years with body mass indexes (BMIs, calculated as weight (kg)/[height (m)]2) of 17.0–35.0 who were at risk for pregnancy, in need of contraception, and were not planning to use condoms were eligible for enrollment. Exclusion criteria were guided by the World Health Organization's Medical Eligibility Criteria11 and the class-labeling of contraceptive steroids.
Using an interactive voice response system, participants were randomly allocated in a 3:1 ratio to receive nomegestrol acetate and 17β-E2 or drospirenone and ethinyl E2 for 13 consecutive 28-day cycles. Within each age class (18–35 years and 36–50 years), randomization was performed using block sizes of four and central allocation in the order of the randomization call. Randomization was not stratified by center.
Study participants received either the investigational product (nomegestrol acetate and 17β-E2: 24 tablets each containing 2.5 mg nomegestrol acetate and 1.5 mg 17β-E2 taken on cycle days 1–24 and four placebo tablets taken on cycle days 25–28) or the comparator (drospirenone and ethinyl E2: 21 tablets each containing 3.0 mg drospirenone and 30 micrograms ethinyl E2 taken on cycle days 1–21 and seven placebo tablets taken on cycle days 22–28).
Women who were not already using hormonal contraception were instructed to begin the trial medication on the first day of menstrual bleeding. Women who began the trial medication within days 2–5 used condoms as a backup during the first 7 days of active trial medication intake. Volunteers switching from another hormonal contraceptive began the trial medication within 7 days after the last active treatment day of the previous method. Volunteers switching from a progestogen-only pill, implant, or intrauterine device began using the trial medication immediately after discontinuing their previous contraceptive method and used a backup method for 7 days.
If an active tablet was taken more than 12 hours late, participants were instructed to take the missed tablet as soon as they remembered. Subsequent tablets were to be taken on schedule, even if study participants were required to take two pills on the same day. When tablets were forgotten, the use of condoms was recommended until 7 days of uninterrupted daily tablet intake occurred (defined differently between the two treatment groups). Nomegestrol acetate and 17β-E2 recipients were allowed to miss one active tablet (or two tablets between days 8 and 17) without being required to use condoms. Drospirenone and ethinyl E2 recipients were only allowed to miss one active tablet during the second week (instructions taken from Yasmin prescribing information). All other active tablet omissions required women to use condoms.
At randomization, participants received an electronic diary and training on its use. This specialized personal digital assistant facilitated secure recording and password-protected transmittal of data into a central database. Participants used e-diaries daily to record tablet intake and vaginal bleeding. Participants input condom use and occurrence of vaginal intercourse once per treatment cycle.
Women visited study centers at screening, randomization, after cycles 3, 6, and 9, end of treatment (cycle 13 or early discontinuation), and posttreatment. Physical examinations, breast examinations, cervical smears, and routine hematology and biochemistry assessments were performed at screening and end of treatment. Blood pressure, body weight (without shoes), and acne were assessed during all treatment visits. Standardized equipment and procedures were used to measure body weight. Serum pregnancy tests were performed at screening, randomization, and end of treatment as well as whenever needed on clinical grounds. If pregnancy was confirmed at any time, follow-up continued until the pregnancy outcome was determined. This included any posttreatment pregnancies with an estimated date of conception within 14 days after discontinuation of trial medication.
This trial was designed in conjunction with a similar trial that took place in Asia, Australia, and Europe.6 The primary goal of these trials was to obtain a combined 95% confidence interval (CI) for the Pearl Index for nomegestrol acetate and 17β-E2 in women age 18–35 years fulfilling the Committee for Medicinal Products for Human Use criterion (ie, the difference between the upper limit of the CI and the point estimate did not exceed 1).12,13 In line with previous trials assessing the efficacy of new oral contraceptives, we assumed that the Pearl Index of nomegestrol acetate and 17β-E2 would have a range of 0.0–2.0. Because the sample size method followed a Poisson distribution, the variance was fixed at 2.0. We assumed that 45% of participants would prematurely discontinue treatment and that 30% of the cycles would not be at risk for pregnancy. The size of the comparator group was determined by power calculations for differences with respect to important safety end points (ie, adverse events and discontinuation rates). These considerations led us to choose a target enrollment of 1,410 participants (nomegestrol acetate and 17β-E2) compared with 470 participants (drospirenone and ethinyl E2) for women age 35 years or younger. For the safety analyses, we also included 330 (nomegestrol acetate and 17β-E2) and 110 (drospirenone and ethinyl E2) additional women age 36–50 years. Overall, 2,320 women were to be randomly assigned to the treatment groups. All women who underwent randomization and received at least one dose of trial medication (all-patients-treated group) were included in the safety analysis.
Contraceptive effectiveness was assessed by recording in-treatment pregnancies from the first day of trial medication intake until 7 days (ie, the 7-day extension window) after last active or placebo tablet intake in the intention-to-treat group. For the purposes of this analysis, and specifically for the life table analyses, the intention-to-treat group included all women who completed at least one cycle (determined from e-diary data). Contraceptive effectiveness was expressed as a Pearl Index, which quantified the number of in-treatment pregnancies per 100 woman-years of exposure (1 woman-year defined as 13 cycles×28 days). The Pearl Index was based on all women in the intention-to-treat group, excluding any cycles in which condoms were always used (determined from e-diary data). Point estimates and corresponding 95% CIs were calculated using the Poisson distribution.14 In addition to the Pearl Index, we estimated pregnancy risk using life table analyses (Kaplan-Meier estimates and corresponding 95% CIs), which were expressed as pregnancy rates per 100 women.
Bleeding patterns were analyzed for all treated women using data from the participant's e-diary. Vaginal bleeding was classified as spotting (requiring one pad or tampon or less per day) or bleeding (requiring greater than one pad or tampon per day). We used a fixed 91-day interval for each reference period,13,15,16 giving a total of four reference periods. The analysis included all treated participants who had at least one evaluable reference period (ie, no more than 2 consecutive days with missing bleeding information occurred within the reference period). One or 2 consecutive days with missing bleeding information were interpolated using bleeding information from the preceding day. We counted the number of bleeding and spotting days for each participant and reference period and the number and mean length of bleeding or spotting episodes. The bleeding patterns between groups were compared using the Wilcoxon rank-sum test stratified by age class (Mantel-Haenszel statistic).17
Bleeding patterns also were assessed by cycle analysis using records from the e-diary. The same interpolation rule for missing data as in the reference period analysis was used. For each treatment group, the expected bleeding period began with the first placebo tablet and lasted for 7 days; the remainder of the cycle constituted the expected nonbleeding period of 21 days.
The two primary bleeding parameters in the cycle analysis were occurrence of unscheduled (breakthrough) bleeding or spotting and the absence of scheduled (withdrawal) bleeding. Unscheduled bleeding or spotting was any bleeding or spotting episode that occurred during the expected nonbleeding days. Scheduled bleeding was any bleeding or spotting episode that began during or continued into the expected bleeding period.
For the primary bleeding parameters, exact binomial, two-sided 95% CIs were calculated per treatment group and cycle. Differences between treatment groups were compared for cycles 2 through 13 using a Mantel-Haenszel χ2 test stratified by age class.17 No comparisons were made during cycle 1 owing to different starting procedures. In cycle 13, we did not complete cycle analysis for absence of scheduled bleeding episodes because the expected bleeding period continued beyond the end of the reporting period (ie, into cycle 14).
Secondary bleeding parameters included occurrence of unscheduled bleeding episodes (ie, breakthrough bleeding episodes with or without spotting days), occurrence of unscheduled spotting episodes (ie, breakthrough bleeding episodes with spotting only), number of unscheduled bleeding or spotting days, and number of scheduled (ie, withdrawal) bleeding or spotting days. For each cycle, the number of unscheduled and scheduled bleeding or spotting days were compared between treatment groups using a Wilcoxon rank-sum test stratified by age class (Mantel-Haenszel statistic).
Compliance was assessed by counting unused pills and using data from e-diaries. Reasons for discontinuation were assessed by the investigators and included a specific question regarding the acceptability of bleeding patterns. Discontinuation from treatment data was analyzed by Kaplan-Meier estimates. To evaluate factors that may have influenced discontinuation rates, data were analyzed (exploratory post hoc analyses) using Cox regression models. Using clinical knowledge, we selected several prognostic factors, including age (18–24, 25–35, and 36–50 years), ethnicity (Asian, African American, white, other), body weight, BMI, smoking (yes or no), and use of a hormonal contraceptive method within 2 months before beginning the trial medication (starter or switcher). The proportional hazards assumption was tested at the .05 significance level by adding a treatment-by-time interaction term to the final model. Several functional forms of the time function were examined (linear, log, quadratic, square root, and inverse). In addition, graphical methods were applied using log(−log) plots of the Kaplan-Meier curves. Graphical and analytical methods did not indicate that the proportional hazards assumption was violated (P>.05 for all methods). The final model included: study treatment, age, ethnicity, smoking status, and starter or switcher status. We also investigated interactions between each covariate and study treatment. Given that no important interactions were found (P>.05), a main effects model could be applied, yielding straightforward interpretation. The hazard ratios and 95% CIs from the Cox regression model were adjusted for all other factors in the model.
At screening and all visits after randomization, clinicians assessed acne in women who received nomegestrol acetate and 17β-E2 or drospirenone and ethinyl E2. Study staff used a 4-point scale (none, mild, moderate, or severe) to assess acne severity according to their own judgment. The changes from baseline to last measurement in acne severity were compared between the treatment groups using a Wilcoxon rank-sum test stratified by baseline values and age class (Mantel-Haenszel statistic). If at any assessment an increase in acne from screening was observed, it was recorded as an adverse event. Safety data were obtained from monitoring laboratory parameters, physical and gynecological examinations at screening and end of treatment, and identification of adverse and serious adverse events. All adverse and serious adverse events from the beginning of study treatment until 28 days after the last active or placebo tablet were considered to be in-treatment events. Throughout the Results and Discussion sections, data are presented first for nomegestrol acetate and 17β-E2 followed by drospirenone and ethinyl E2.
In June 2006, the first participant was enrolled in the study. The study was completed in July 2008. Baseline demographic characteristics of participants (including age, BMI, and smoking status) were similar between treatment groups (Table 1). By design, more than 80% of women in both treatment groups were age 18–35 years. Most women in the nomegestrol acetate and 17β-E2 and the drospirenone and ethinyl E2 groups (87.8% and 85.3%, respectively) had BMIs of 30 or less. Of participants who underwent randomization (n=2,281) and were assigned to receive nomegestrol acetate and 17β-E2 (n=1,710) or drospirenone and ethinyl E2 (n=571), 1,332 participants completed 1 year of study treatment (Fig. 1). Approximately 41% and 38% of recipients in the respective groups discontinued treatment before the end of the trial.
The total exposure used for the calculation of the pregnancy rate in women age 18–35 years was 12,296 cycles (946 woman-years) and 4,135 cycles (318 woman-years) for each respective treatment group (Table 2). The Pearl Index in women age 18–35 years was 1.27 (95% CI, 0.66–2.22) and 1.89 (95% CI, 0.69–4.11) for the nomegestrol acetate and 17β-E2 and drospirenone and ethinyl E2 groups, respectively. Although contraceptive efficacy of nomegestrol acetate and 17β-E2 was slightly better compared with drospirenone and ethinyl E2, the difference between treatment groups was not statistically significant. Life table analyses were consistent with Pearl Index estimations, including treatment group comparisons (log-rank test and Wilcoxon test from standard life table analyses, P=.44 and P=.47, respectively).
From reference periods 1–4, the mean overall number of bleeding days continuously decreased among women who received nomegestrol acetate and 17β-E2 (from 5.9 to 4.1 days). In the drospirenone and ethinyl E2 group, the mean overall number of bleeding days steadily increased from reference period 1 (9.8 days) to reference period 4 (11.6 days). The mean number of bleeding days was substantially lower for all reference periods in the nomegestrol acetate and 17β-E2 group (P<.001 compared with drospirenone and ethinyl E2). In the nomegestrol acetate and 17β-E2 group, the mean number of spotting days decreased from 8.9 days (reference period 1) to 5.4 days (reference period 4). In the drospirenone and ethinyl E2 group, the mean number of spotting days remained unchanged (from 7.9 to 7.7 days). The mean number of spotting days was significantly lower with nomegestrol acetate and 17β-E2 for reference periods 3 (P=.031) and 4 (P<.001). Treatment groups were similar with regard to spotting in the first two reference periods. Women in the nomegestrol acetate and 17β-E2 group had fewer bleeding or spotting days from reference period 1 onward (Fig. 2A), which was predominantly attributable to the reduced number of bleeding days. For each reference period, the median number of the mean length of bleeding or spotting episodes was approximately 1 day shorter in the nomegestrol acetate and 17β-E2 group (Fig. 2B). The median number of bleeding or spotting days and the median number of the mean length of bleeding or spotting episodes were all significantly lower and shorter with nomegestrol acetate and 17β-E2 for all reference periods (P<.001).
In the per-cycle analyses, the median number of unscheduled bleeding or spotting days (for women who experienced unscheduled bleeding episodes) was comparable between treatment groups (Fig. 3A), although differences in the number of unscheduled bleeding or spotting days were noted for some cycles (cycles 2, 5, 8, and 10; P<.05 compared with drospirenone and ethinyl E2). The median unscheduled bleeding or spotting days fluctuated between 2 and 3 days in both treatment groups. The incidence of unscheduled bleeding or spotting episodes was higher in the nomegestrol acetate and 17β-E2 group (Fig. 3B), especially during the first six cycles. The median scheduled bleeding or spotting days per cycle (for women who experienced scheduled bleeding episodes) was lower in the nomegestrol acetate and 17β-E2 group (Fig. 4A) (P<.001 for all cycles). The absence of scheduled bleeding occurred more frequently in the nomegestrol acetate and 17β-E2 group in all cycles (cycles 2–12; P<.001; Fig. 4B). By the end of the trial, scheduled bleeding episodes were shorter and lighter and were sometimes absent altogether in women who received nomegestrol acetate and 17β-E2. The all-patients-treated group (women age 18–50 years) used for safety analyses included 2,220 participants (1,666 and 554 women in each respective group).
The presence of acne at baseline was similar between treatment groups (33.3% and 33.8%, respectively) and decreased over time in both groups (Fig. 5). Most participants who did not have acne at baseline also were free from acne at last measurement (87.6% and 95.8% for each respective treatment group). There were more new cases of acne in women who received nomegestrol acetate and 17β-E2 (12.4%) than in those who received drospirenone and ethinyl E2 (4.2%). In both groups, women with acne at baseline were much more likely to show improvement (54% compared with 63%) at last measurement than to have acne worsen (6% compared with 5%) or remain unchanged (40% compared with 32%). Differences in change in acne from baseline to last measurement were in favor of drospirenone and ethinyl E2 (P<.001).
One or more adverse event(s) were reported in 70.6% (n=1,177) and 62.5% (n=346) of participants in each respective treatment group. In both groups, 48.8% (n=813) and 36.3% (n=201) of women reported adverse events that were judged by the investigators to be related to the study treatment. The most frequently reported treatment-related adverse events (defined as having an incidence of 5% or greater) for nomegestrol acetate and 17β-E2 and drospirenone and ethinyl E2, respectively, were acne (16.4% compared with 8.7%), weight gain (9.5% compared with 5.2%), irregular withdrawal bleeding (9.1% compared with 0.5%), and metrorrhagia (5.8% compared with 2.7%).
During the in-treatment period, serious adverse events were reported by 35 women (1.8% of nomegestrol acetate and 17β-E2 recipients; 0.9% of drospirenone and ethinyl E2 recipients). The study investigators judged that serious adverse events in five women were related to study treatments (two cases of cholelithiasis and one case each of cholecystitis, optic neuritis, and migraine, all in women using nomegestrol acetate and 17β-E2). One woman in the nomegestrol acetate and 17β-E2 group died in a traffic accident. As per convention, this participant was considered treated for all analyses, but it was unknown whether she had begun to take the study medication or if she experienced an antecedent event before the traffic accident.
Reasons for discontinuation from study treatment are shown in Figure 1. Unacceptable bleeding (ie, irregular withdrawal bleeding) accounted for 3.8% and 1.8% of discontinuations in each respective treatment group (included in percentage of women in each treatment group who discontinued owing to an adverse event). Other adverse events, including acne and weight gain, accounted for 13.5% and 8.3% of discontinuations, respectively. Discontinuation owing to irregular withdrawal bleeding occurred gradually over time among women who received nomegestrol acetate and 17β-E2. Differences between the groups were significant for unacceptable bleeding and other adverse events (P<.023 and P<.003, respectively). In the multivariable Cox proportional hazards model, age, ethnicity, smoking status, and starter or switcher status had a significant influence on discontinuation from treatment (P<.001 for each factor). Results of the comparisons of these factors are summarized in Table 3. Starters, younger women, and those who smoked also were more likely to discontinue treatment. Neither body weight nor BMI nor obesity (BMI of 30 or higher) was associated with discontinuation.
Laboratory and blood pressure measurements showed no remarkable changes in values from baseline in either treatment group. Weight increased from 65.8 kg at baseline to 66.7 kg (median 1.0 kg) and 65.8 kg to 66.0 kg (median 0.2 kg) at last measurement in the two groups, respectively. The modest increase in weight from baseline until last measurement was significant in both groups (P=.001).
Results of this study and a similar trial6 show that nomegestrol acetate and 17β-E2 provides robust contraceptive efficacy and shorter, lighter withdrawal bleeding compared with drospirenone and ethinyl E2. This new 24–4 regimen combined oral contraceptive was well tolerated and had an acceptable safety profile. In both trials, the 1-year cumulative pregnancy rates were lower among women who received nomegestrol acetate and 17β-E2 than those who received drospirenone and ethinyl E2, although this difference did not achieve statistical significance. The contraceptive efficacy of nomegestrol acetate and 17β-E2 was achieved despite applying less stringent requirements for using condoms as a backup method when women missed tablets than those requirements applied to the comparator. Although use of the Pearl Index is popular, this method has limitations, including the assumption of an underlying constant pregnancy risk over time and the independence of all exposure cycles. If the Pearl Index is used, life table analysis should also be used to validate and reaffirm contraceptive effectiveness.
The high contraceptive efficacy of nomegestrol acetate and 17β-E2 is consistent with the ovarian suppression previously observed4 and also may be related to the shorter hormone-free interval.18 The longer terminal elimination half-life (approximately 46 hours) of nomegestrol acetate in comparison with drospirenone (31.1–32.5 hours) also may have contributed to the trend toward lower Pearl Index values.7,18,19
In comparison with drospirenone and ethinyl E2, nomegestrol acetate and 17β-E2 was associated with a vaginal bleeding pattern characterized by fewer scheduled bleeding episodes, which were of shorter duration and lighter intensity. Absence of scheduled bleeding may be a feature inherent to combined oral contraceptive regimens with shorter hormone-free intervals20 and may have been accentuated by the long elimination half-life of nomegestrol acetate.7 Absence of withdrawal bleeding is becoming more acceptable,20 but it was not anticipated at the start of this trial. Consequently, investigators and women were not counseled on the unique bleeding profile of nomegestrol acetate and 17β-E2, which could have contributed to lower acceptance (as reflected in the adverse event and discontinuation data) among trial participants, some of whom may have equated absence of bleeding with unintended pregnancy. We hypothesize that anticipatory guidance regarding the paucity of vaginal bleeding and high contraceptive effectiveness will result in greater recipient satisfaction with nomegestrol acetate and 17β-E2.
Although both oral contraceptives reduced acne in women with acne at baseline, this effect was more pronounced in women who received drospirenone and ethinyl E2. Frequent clinician assessments may have helped to increase acne awareness, increasing the frequency of acne reporting.
Nomegestrol acetate and 17β-E2 recipients experienced a small increase in mean body weight, which was slightly larger than the results observed in drospirenone and ethinyl E2 recipients. This difference may be explained in part by the antimineralocorticoid action of drospirenone, which results in some loss of body water during the first months of treatment.21,22 Overall, there is no causal relationship between combined oral contraceptive use and body weight changes.23,24 Women of reproductive age who use nonhormonal contraception or abstain from using contraceptives altogether are known to gain weight,25–27 underscoring that weight gain over time is common irrespective of contraceptive choice.
In the twin study that evaluated the contraceptive efficacy of nomegestrol acetate and 17β-E2 in Europe, Australia, and Asia, the discontinuation rate was 28.2%.6 The discontinuation rate in the present trial was comparable to rates observed for other contraceptives evaluated in the United States.28,29 A U.S. observational study also found that among combined oral contraceptive initiators, nearly 60% of these women discontinued use within 6 months.28
In summary, nomegestrol acetate and 17β-E2 was well tolerated and had an acceptable adverse event profile. Users more frequently reported acne, irregular withdrawal bleeding, weight gain, and sometimes discontinued secondary to these conditions. These observed differences between the groups may reflect differences in the pharmacological properties of the hormones and treatment regimens. This comparative trial found that nomegestrol acetate and 17β-E2 provides good contraceptive effectiveness over 1 year and has a bleeding profile characteristic of combined oral contraceptives with a 24–4 dosing regimen.
2. Astedt B, Jeppsson S, Liedholm P, Rannevik G, Svanberg L. Clinical trial of a new oral contraceptive pill containing the natural oestrogen 17beta-oestradiol. Br J Obstet Gynaecol 1979;86:732–6.
3. Serup J, Bostofte E, Larsen S, Westergaard J, Lebech PE. Natural oestrogens for oral contraception. Lancet 1979;2:471–2.
4. Duijkers IJM, Klipping C, Grob P, Korver T. Effects of a monophasic combined oral contraceptive containing nomegestrol acetate and 17β-oestradiol on ovarian function in comparison to a monophasic combined oral contraceptive containing drospirenone and ethinylestradiol. Eur J Contracept Reprod Health Care 2010;15:314–25.
5. Gaussem P, Alhenc-Gelas M, Thomas JL, Bachelot-Loza C, Remones V, Dali Ali F, et al.. Haemostatic effects of a new combined oral contraceptive, nomegestrol acetate/17β-estradiol, compared with those of levonorgestrel/ethinyl estradiol: a double-blind, randomised study. Thromb Haemost 2011;105:560–7.
6. Mansour D, Verhoeven C, Sommer W, Weisberg E, Taneepanichskul S, Melis GB, et al.. Efficacy and tolerability of a monophasic combined oral contraceptive containing nomegestrol acetate and 17β-oestradiol in a 24/4 regimen, in comparison to an oral contraceptive containing ethinylestradiol and drospirenone in a 21/7 regimen. Eur J Contracept Reprod Health Care 2011;16:430–43.
7. Lello S. Nomegestrol acetate: pharmacology, safety profile and therapeutic efficacy. Drugs 2010;70:541–59.
8. van Diepen HA, Lam TW, Kuil CW. Nomegestrol acetate: steroid receptor transactivation profile in Chinese hamster ovary cells and ovulation inhibition in rat and monkey. Contraception 2011;84:199–204.
9. Ågren UM, Anttila M, Mäenpää-Liukko K, Rantala ML, Rautiainen H, Sommer W, et al.. Effects of a monophasic combined oral contraceptive containing nomegestrol acetate and 17β-oestradiol in comparison to one containing levonorgestrel and ethinylestradiol on markers of endocrine function. Eur J Contracept Reprod Health Care 2011;16:458–67.
10. Ågren UM, Anttila M, Mäenpää-Liukko K, Rantala ML, Rautiainen H, Sommer W, et al.. Effects of a monophasic combined oral contraceptive containing nomegestrol acetate and 17β-oestradiol compared with one containing levonorgestrel and ethinylestradiol on haemostasis, lipids and carbohydrate metabolism. Eur J Contracept Reprod Health Care 2011;16:444–57.
11. World Health Organization; Reproductive Health and Research. Medical eligibility criteria for contraceptive use. 3rd ed. Geneva (Switzerland): World Health Organization; 2004.
12. Benda N, Gerlinger C, van der Meulen EA, Endrikat J. Sample size calculation for clinical studies on the efficacy of a new contraceptive method. Biom J 2004;46:141–50.
13. European Medicines Agency; Committee for Medicinal Products for Human Use. Guideline on clinical investigation of steroid contraceptives in women. London (UK): European Medicines Agency; 2005.
14. Gerlinger C, Endrikat J, van der Meulen EA, Dieben TOM, Düsterberg B. Recommendation for confidence interval and sample size calculation for the Pearl Index. Eur J Contracept Reprod Health Care 2003;8:87–92.
15. Belsey EM, Farley TM. The analysis of menstrual bleeding patterns: a review. Contraception 1988;38:129–56.
16. Belsey EM, Machin D, d'Arcangues C. The analysis of vaginal bleeding patterns induced by fertility regulating methods. World Health Organization Special Programme of Research, Development and Research Training in Human Reproduction. Contraception 1986;34:253–60.
17. Der G, Everitt BS. Statistical analysis of medical data using SAS. Boca Raton (FL): Chapman & Hall/CRC, Taylor & Francis Group, LLC; 2006.
18. Dinger J, Minh TD, Buttmann N, Bardenheuer K. Effectiveness of oral contraceptive pills in a large US cohort comparing progestogen and regimen. Obstet Gynecol 2011;117:33–40.
19. Stanczyk FZ. Pharmacokinetics and potency of progestins used for hormone replacement therapy and contraception. Rev Endocr Metab Disord 2002;3:211–24.
20. Read CM. New regimens with combined oral contraceptive pills—moving away from traditional 21/7 cycles. Eur J Contracept Reprod Health Care 2010;15:S32–41.
21. Fruzzetti F, Lazzarini V, Ricci C, Quirici B, Gambacciani M, Paoletti AM, et al.. Effect of an oral contraceptive containing 30 μg ethinylestradiol plus 3 mg drospirenone on body composition of young women affected by premenstrual syndrome with symptoms of water retention. Contraception 2007;76:190–4.
22. Guang-Sheng F, Mei-Lu B, Li-Nan C, Xiao-Ming C, Zi-Rong H, Zi-Yan H, et al.. Efficacy and safety of the combined oral contraceptive ethinylestradiol/drospirenone (Yasmin®) in healthy Chinese women: a randomized, open-label, controlled, multicentre trial. Clin Drug Investig 2010;30:387–96.
23. Gallo MF, Lopez LM, Grimes DA, Schulz KF, Helmerhorst FM. Combination contraceptives: effects on weight. The Cochrane Database of Systematic Reviews 2011, Issue 9. Art. No.: CD003987. DOI: 10.1002/14651858.CD003987.pub4.
24. Lindh I, Ellström AA, Milsom I. The long-term influence of combined oral contraceptives on body weight. Hum Reprod 2011;26:1917–24.
25. Coney P, Washenik K, Langley RG, DiGiovanna JJ, Harrison DD. Weight change and adverse event incidence with a low-dose oral contraceptive: two randomized, placebo-controlled trials. Contraception 2001;63:297–302.
26. Thiboutot D, Archer DF, Lemay A, Washenik K, Roberts J, Harrison DD. A randomized, controlled trial of a low-dose contraceptive containing 20 μg of ethinyl estradiol and 100 μg of levonorgestrel for acne treatment. Fertil Steril 2001;76:461–8.
27. Hassan DF, Petta CA, Aldrighi JM, Bahamondes L, Perrotti M. Weight variation in a cohort of women using copper IUD for contraception. Contraception 2003;68:27–30.
28. Westhoff CL, Heartwell S, Edwards S, Zieman M, Stuart G, Cwiak C, et al.. Oral contraceptive discontinuation: do side effects matter? Am J Obstet Gynecol 2007;196:412.e1–6; discussion 412.e6–7.
29. Dieben TO, Roumen FJ, Apter D. Efficacy, cycle control, and user acceptability of a novel combined contraceptive vaginal ring. Obstet Gynecol 2002;100:585–93.