Combined hormonal oral contraceptive pills (OCPs) have been used widely since 1960. Although effective in preventing pregnancy, safety concerns about high doses of progestin and estrogen in early formulations remained. Common side effects of such OCPs include breast tenderness, headache, migraine, nausea, nervousness, weight gain, and hypertension1; the estrogen component, specifically, is associated with cerebrovascular complications, thromboembolic incidents, and myocardial infarction.2 Therefore, estrogen doses have decreased to 20–35 micrograms per pill for currently used OCPs without affecting overall efficacy.3 Oral contraceptive pills often are given in 21-day cycles of active pills followed by 7 days of placebo or no pills.
Inconsistent compliance and discontinuation are common among OCP users,4,5 often as a result of adverse effects and bleeding irregularities,6 and are associated strongly with pregnancy.5,7 Lower-estrogen OCPs are associated with higher rates of breakthrough bleeding and spotting than higher-estrogen formulations.2 An extended duration of active hormone treatment for more than 21 days may decrease breakthrough bleeding and discontinuations associated with lower-estrogen OCPs8 as well as improve efficacy.7
This study investigated the efficacy in preventing pregnancy, effects on intracyclic bleeding, and safety of an ultra low-dose OCP (1.0 mg norethindrone acetate and 10 micrograms ethinyl estradiol [E2]) administered for 24 days, followed by 2 days 10 micrograms ethinyl E2 and another 2 days of inactive tablet (28-day cycle), thereby halving the daily estrogen dose from 20 micrograms to 10 micrograms and extending estrogen dosing beyond 24 days. This regimen may have the potential to reduce medical side effects and improve failure rates and bleeding irregularities.
PATIENTS AND METHODS
This was a phase III, open-label, uncontrolled, multicenter study conducted at 68 centers in 21 states across the United States. Each site recruited locally from their practices and by advertising. The primary objective was to assess the efficacy of an ultra low-dose OCP regimen in preventing pregnancy. Secondary objectives included assessment of intracyclic bleeding, safety, and tolerability. Pearl Index and life-table methods were used to assess the pregnancy rate. Bleeding occurrences were monitored through a daily diary. The study conformed to the International Conference on Harmonization, Harmonized Tripartite Guideline for Good Clinical Practice, and the Declaration of Helsinki and amendments. The protocol, amendments, and informed consent documents were reviewed and approved by appropriate investigational review boards (including Eastern Virginia Medical School, Medical University of South Carolina, Oregon Health and Science University, Schulman Associates institutional review board, University of Louisville, and Western institutional review board). All participants were required to provide continuing, written consent.
The study population comprised heterosexually active women aged 18–45 years with a body mass index (BMI, calculated as weight (kg)/[height (m)]2) of 35 or less, a negative serum pregnancy test, and a willingness to use the study drug as the only method of contraception. At enrollment, eligible participants had regular menstrual cycles with a usual length of 21–35 days and a variability of ±3 days (recently postpartum or postabortion participants had two or more normal cycles before enrollment). Exclusion criteria were: 1) use of progestational implants, progestin, estrogen or estrogen and progestational injectable drug therapy within 9 months, or intrauterine contraception within 3 months; 2) an abnormal Pap test suggestive of a low-grade squamous intraepithelial lesion or a more serious problem; 3) an untreated Chlamydia trachomatis infection; 4) breastfeeding; 5) any abnormal findings or condition on medical history, screening, physical and gynecologic examination (including abnormal baseline laboratory values), known or suspected malignant disease, cardiovascular disease, uncontrolled hypertension, thrombophlebitis or thromboembolic disorder, or undergoing treatment with anticoagulants; 6) a history of drug addiction or alcohol abuse (within 2 years); or 7) smoked 15 or more cigarettes per day if aged 35 years or older.
The treatment regimen for each 28-day pill cycle was one oral tablet containing 1.0 mg norethindrone acetate and 10 micrograms ethinyl E2 daily for 24 days followed by one oral 10-microgram ethinyl E2 tablet daily for 2 days followed by one ferrous fumarate tablet on each of the final 2 days. Participants were asked to take the study drug at approximately the same time each day and treatment was scheduled to continue for 13 28-day cycles. Women who had been using an OCP immediately before enrolling in this study (defined as “switchers”) started the study drug on the same day that they would have resumed the active formulation of the previous product (day 1). If unscheduled spotting or bleeding occurred during treatment, the regimen was to be continued without interruption. Women who had not been using an OCP before the study began taking the study drug on the first day of menstrual flow (day 1).
The primary end point and efficacy outcome was the incidence of pregnancy based on the Pearl Index (defined as the number of pregnancies per 100 women-years of treatment) in the group aged 35 years or younger based on all at-risk cycles in which no other contraceptive method was used. Also presented are the Pearl Indices for the overall (all participants regardless of age) and older cohorts (36 years of age or older). The estimated date of conception was determined from ultrasonography and was compared with the start and stop dates for study medication. A pregnancy was counted in the efficacy analysis if the estimated date of conception occurred 14 days or more after the first date of study medication and up to and including 14 days after the last day of study medication.
Secondary end points were intracyclic bleeding or spotting, treatment compliance, safety, and tolerability during treatment. A bleeding episode was defined as 1 or more day of bleeding or spotting, which could be consecutive or interrupted once or more by no more than 1 day without bleeding or spotting. A withdrawal bleeding episode was defined as the first bleeding episode starting 4 days before the last day of active drug intake during a treatment cycle and 3 days after beginning treatment in the next treatment cycle. All other bleeding (spotting) days and episodes were counted as intracyclic bleeding or spotting days and episodes. The presence and intensity of withdrawal bleeding was scored in a diary as: 0=none, 1=lighter than normal, 2=normal, or 3=heavier than usual. Whether bleeding required the use of sanitary protection (other than panty liners) was also recorded. Light bleeding not requiring sanitary protection was classified as spotting. The duration of a bleeding (spotting) episode was defined as the number of days from the first to the last day of the episode, inclusive, including single, isolated, and bleed-free (spot-free) days. The intensity of the bleeding (spotting) episode was defined as the maximum intensity of the bleeding (spotting) days, in which 1=light, 2=normal, and 3=heavy. Participants also recorded any other contraceptive methods used in addition to or instead of the study drug. Treatment compliance was assessed using each participant's diary and returned blister packs.
Pregnancy status, diaries, and vital signs were assessed at eight study visits including baseline; interim visits at the end of cycles 1, 3, 5, 7, 9, and 11; and a final visit after cycle 13. Follow-up after the study was carried out through a participant-completed questionnaire. Weight, gynecologic assessments, and laboratory tests were performed at baseline and at the final visit. Adverse events were recorded by severity (mild, moderate, or severe) at each visit using Medical Dictionary for Regulatory Activities coding.
The statistical analysis of efficacy included all participants who received treatment and who were evaluated for pregnancy at least once (modified intent-to-treat population). The Pearl Index and 95% confidence intervals for all participants, regardless of age, were calculated based on all at-risk cycles when no other contraceptive method was used. In the analysis of intracyclic bleeding or spotting, single missing diary days were imputed using the maximum bleeding intensities of the day before and after the missing day. No imputations were made for 2 or more consecutive missing diary days (these days were not evaluable). Any cycle with fewer than 14 evaluable diary days was excluded from the analysis. All bleeding parameters were summarized using appropriate descriptive statistics by reference period and cycle, by age cohort, by user status (switcher compared with new start), and overall.
A total of 2,235 potential participants were screened and 1,683 were enrolled. Participant disposition is depicted in Figure 1. A total of 1,660 enrolled and treated participants (all treated population) took at least one dose of the study medication; 23 enrolled participants did not take any medication and returned all of the study medication unused. The discontinuation rate was 41.7% (692/1,660). Participants discontinued from the all treated population for the following reasons: 227 (13.7%) were lost to follow-up, 177 (10.7%) withdrew as a result of adverse events, 147 (8.9%) withdrew consent, 96 (5.8%) withdrew for “other reasons,” 20 (1.2%) had protocol violations, and 25 (1.5%) withdrew as a result of lack of efficacy (pregnancy; one pregnancy was diagnosed after the participant had finished the study). The modified intent-to-treat population (n=1,582) was defined as the subset of the all treated population who were evaluated for pregnancy, whether positive or negative, at least once after beginning the study medication. The completed participants population (n=968) was defined as the subset of the modified intent-to-treat population who completed at least 360 days of treatment based on diary reports.
Participant baseline characteristics are summarized in Table 1. Most participants (81.9%) in the modified intent-to-treat population were aged 18–35 years; 51.0% had not been using OCPs, intravaginal devices, or transdermal contraceptive methods immediately before enrolling; and most (82%) had a BMI less than 30 with an average BMI of 25.2 (range 15.7–38.3).
Participants took study medication for a mean of 272.4 days and 9.9 cycles (all treated population). A total of 105 (6.7%) participants with 1,133 (7.5%) cycles were excluded from the analysis. The average number of pills taken per cycle was 25.3 (range 0–28). The mean exposure to study medication in the modified intent-to-treat population was 273.8 days (median 356 days), the mean number of cycles treated was 10.4 (median 13 cycles), and the average number of pills taken per cycle was 25.4 out of a possible 28, of which two were considered placebo.
The pregnancy outcomes of the modified intent-to-treat population (n=1,582) are presented in Table 2. Twenty-seven women had no cycles evaluable for pregnancy because they used alternative contraceptive methods in all cycles. Of the remaining 1,555 women, 26 became pregnant during 15,596 at-risk cycles while using the ultra low-dose OCP regimen. The Pearl Index was 2.2 and the cumulative pregnancy rate was 2.1 for the overall population. The Pearl Indices were 2.6 and 0.8 and the cumulative rates were 2.5 and 0.6, respectively, for women aged 18–35 years and 36 years or older, respectively. Participants whose pregnancies were not counted in the calculation of the Pearl Index included nine found to be pregnant after enrollment but before starting the study drug, four found to be pregnant more than 14 days after taking the last active pill, and one found to be pregnant less than 14 days after taking the first active pill. The Pearl Indices were 2.2 and 1.9 for women with a BMI less than 30 and 30 or more, respectively (Table 2).
A total of 1% of bleeding diaries required imputation. In the modified intent-to-treat population, an average of 2.6 days of intracyclic bleeding or spotting per cycle was experienced (during cycles 2–13). The mean number of days of intracyclic bleeding or spotting decreased from 3.2 days in cycle 2 to 1.8 days in cycle 13 (Fig. 2A). The mean maximum intensity score (ie, the mean across each participant's maximum intensity in the cycle or cycles) for intracyclic bleeding or spotting episodes was 1.6 for cycle 2 to cycle 13 in the modified intent-to-treat population (Fig. 2B). Intracyclic bleeding or spotting occurred in 52.7% of participants during cycle 2 and fell below 40% in cycle 8; the lowest incidence (36.4%) was in cycle 13 (Fig. 2C). Overall, the incidence of intracyclic bleeding or spotting tended to be higher in new users than in switchers but gradually decreased after cycle 2 or 3 in both subpopulations. Intracyclic bleeding or spotting was also more frequent in younger (18–35 years) than older women (36 years or older). The number of days with intracyclic bleeding or spotting decreased during the study for new users and switchers as well as for the younger and older subpopulations. Mean maximum intensity scores tended to be higher in younger than in older participants.
The percentage of participants with withdrawal bleeding decreased during the study. Withdrawal bleeding was highest during cycle 1 (43.3%), decreasing to 22.4% during cycle 13. The mean median duration of withdrawal bleeding was 4.0 days for cycles 2–13 (Fig. 3A). The overall mean median intensity of withdrawal bleeding for cycles 2–13 was 1.5 for the modified intent-to-treat population. The duration of withdrawal bleeding for cycles 2–13 tended to be longer in new users (4.3 days) than in switchers (3.7 days). No trends with respect to age were noted.
Total bleeding days per cycle was low with a mean of 3.8 days per cycle for cycles 2–13 for the modified intent-to-treat population (Fig. 3B). The total number of bleeding days per cycle tended to decrease, ranging from 4.6 days per cycle in cycle 2 to 2.7 days per cycle in cycle 12. The total number of bleeding days was greater in new users (4.5 days) compared with switchers (3.2 days).
A complete absence of all bleeding or spotting (withdrawal and unscheduled) during a cycle was common. The incidence was 31.6% in cycle 1 (modified intent-to-treat population), increasing to 49.1% in cycle 13. The incidence of absence of bleeding was higher in switchers than in new users and in older compared with younger participants.
One or more treatment-emergent adverse events were reported by 1,064 participants (64.1%), most commonly headache, upper respiratory tract infections, and sinusitis (Table 3). Most adverse events were mild or moderate in severity. Adverse events considered by the investigators to be treatment-related were reported in 442 participants (26.6%); the most frequently reported were headache (4.8%), metrorrhagia (3.3%), nausea (3.2%), and breast tenderness (3.0%). Treatment-emergent adverse events led to discontinuation of 172 participants (10.4%). Five participants discontinued as a result of adverse events that started before the initiation of drug treatment. The adverse events most commonly leading to discontinuation included metrorrhagia (n=46), menstruation irregularities (n=12), headache (n=10), and mood swings (n=9). A total of 18 serious adverse events were reported in 15 participants (0.9%) (Table 3).
The lower ethinyl E2 dose in the ultra low-dose OCP regimen has the potential of reducing risks and medical side effects associated with higher estrogen doses; however, this could lead to a higher failure rate and an increase in breakthrough bleeding, which in turn could lead to discontinuation and pregnancy. Extending the duration of active pills may balance these risks by lowering the failure rate in decreasing bleeding irregularities.
Our findings demonstrate that this ultra low-dose OCP regimen is effective in preventing pregnancy. Through 13 cycles of treatment, there were 26 pregnancies during 15,596 at-risk cycles with a Pearl Index of 2.2 and a cumulative pregnancy rate of 2.1 for the overall population. The Pearl Index was 2.6 in participants aged 18–35 years and 0.8 in those aged 36 years or older, reflecting the decline in female fecundity with age. The observed efficacy is achieved by balancing the lower estrogen dose with a longer duration of active hormone in this unique regimen.
The inclusion of women aged 36–45 years is noteworthy. Oral contraceptive pills often are not studied in this age group because of declining fertility and a high incidence of spontaneous miscarriage and elective pregnancy termination for unanticipated pregnancies.9 The ultra low-dose OCP regimen was effective in this age group with a Pearl Index of 0.8. This regimen may be particularly useful in older women (36–45 years) considering that age older than 35 years has been shown to influence the choice of low-dose OCPs.10 Women with a BMI 35 or less were included; pregnancy rates in nonobese (BMI less than 30) and obese (BMI 30 or more) participants were similar, indicating that body weight does not affect the pregnancy rate.11
The incidence of intracyclic bleeding or spotting decreased during treatment. The high ratio of progestin to estrogen resulting in a progestin dominant effect on the endometrium could contribute to endometrial vascular fragility, similar to that found with progestin-only implants.12 This finding suggests that either lower doses of ethinyl E2 or the higher amount of norethindrone acetate contributes to a vascular dysfunction during the early cycles of pill use.
The intensity of intracyclic bleeding or spotting was light to normal, the incidences decreased during treatment, and the discontinuation rates were not unusual. The mean number of intracyclic bleeding or spotting episodes, duration of bleeding episodes, and mean maximum intensity of episodes were typically higher in new users than in switchers. In addition, the incidence and mean number of intracyclic bleeding or spotting episodes and duration and mean maximum intensity of bleeding episodes were higher in younger (18–35 years) than in older (36 years or older) participants.
Withdrawal bleeding was infrequent initially (43.3%) in contrast to the higher rate of intracyclic bleeding or spotting during treatment. Withdrawal bleeding was highest during cycle 1 but decreased thereafter. As a result of the definitions of withdrawal and intracyclic bleeding or spotting, a portion of the relatively low percentage of withdrawal bleeding in initial cycles may be reflected in the incidence of intracyclic bleeding or spotting. Total bleeding days per cycle were low and decreased throughout the study, probably as a result of the incidence of absence of bleeding (intracyclic and withdrawal), which increased throughout the study. New users tended to have less absence of bleeding than switchers. The low level of withdrawal bleeding and the incidences of absence of any bleeding might be the result of the short duration (2 days) of estrogen absence in this regimen.
This ultra low-dose OCP regimen was generally well tolerated. Treatment-related adverse events occurred in 26.6% of participants and were typically mild in intensity. Treatment discontinuation as a result of a treatment-emergent adverse event was low. Most of the reported adverse events (mild, moderate, severe, and those leading to treatment discontinuation) were commonly associated with OCPs, including headache, metrorrhagia, irregular menstruation, mood swings, weight fluctuation, amenorrhea, and acne.1 Discontinuations resulting from bleeding anomalies were infrequent.
The results of this study demonstrate that this ultra low-dose OCP regimen has an efficacy that is comparable with that reported for other low-dose OCPs. The overall safety and tolerability profile of the regimen was acceptable with an adverse event profile and discontinuation rate similar to that reported for other low-dose OCPs.2,8,13 These efficacy, safety, and tolerability results were achieved with a regimen that included, in addition to 1.0 mg norethindrone acetate for 24 days, a very low 10-microgram daily dose of ethinyl E2 during the 26 days of each 28-day cycle.