Efficacy of cyclic and extended regimens of ethinylestradiol 0.02 mg ‐levonorgestrel 0.09 mg for dysmenorrhea: A placebo‐controlled, double‐blind, randomized trial

Abstract Purpose We aimed to evaluate the efficacy and safety of 28‐day Cyclic and 84‐day Extended regimens of NPC‐16 (ethinylestradiol 0.02 mg plus levonorgestrel 0.09 mg) in patients with dysmenorrhea. Methods This was a placebo‐controlled, double‐blind, randomized trial conducted in Japan. A total of 251 primary and secondary dysmenorrhea patients were randomly assigned to the NPC‐16‐Cyclic group, NPC‐16‐Extended group, or the Placebo group. The primary end point was a comparison of the efficacy and safety of the Cyclic and Extended NPC‐16 regimen for the treatment of dysmenorrhea relative to the Placebo. Main findings Significantly greater reductions in total dysmenorrhea score and visual analog scale score were observed in the Cyclic and Extended groups compared with the Placebo group. Compared with the Cyclic regimen as a secondary end point, the Extended regimen exhibited greater efficacy in the treatment of dysmenorrhea over the course of the study period, particularly in patients with severe dysmenorrhea. The incidence of adverse drug reactions (ADRs) was significantly higher in the Cyclic and Extended groups than in the Placebo group. Conclusion The Cyclic and Extended regimens of NPC‐16 significantly reduced dysmenorrhea severity compared to placebo. The Extended regimen was superior to cyclic regimen in reducing the dysmenorrhea.


| INTRODUC TI ON
Dysmenorrhea is a common gynecologic problem, reportedly affecting almost half of menstruating women. Of those affected, approximately 5%-15% experience pain of such severity that it limits their ability to work or study, resulting in periods of workplaces or school absence. 1,2 Dysmenorrhea is also a major symptom of endometriosis, which affects 10% of reproductive-age women. Severe, persistent dysmenorrhea at the time of adolescence is a diagnostic marker of adult endometriosis. 3 It was recently proposed that women with severe primary dysmenorrhea should receive early treatment with hormonal drugs before a definite diagnosis of endometriosis by surgical laparoscopy is made, as "look and treat" surgery does not always result in a favorable outcome in terms of the patient's endometriosis life. 4 Such potentially unfavorable outcomes are due to the high recurrence rate after surgery and damage to the ovarian reserve after endometrioma surgery. 5 The use of combined oral contraceptives (OCs) is beneficial for the treatment of dysmenorrhea. We previously conducted randomized controlled trials (RCTs) of the estrogen/progestin combination product IKH-01 (ethinylestradiol [EE] 0.035 mg and norethisterone 1 mg) for the treatment of primary dysmenorrhea and endometriosisassociated dysmenorrhea, and NPC-01 (EE 0.02 mg and norethisterone 1 mg) for the treatment of dysmenorrhea. [6][7][8] The results of these RCTs demonstrated that low-dose OCs are efficacious for controlling pain in patients with either primary or secondary dysmenorrhea.
Compared with a 28-day cyclic regimen, extended regimens significantly reduce the duration of menstrual pain in women using the contraceptives. 9 In our recent RCT conducted in Japan, a lowdose OC (EE 0.02 mg and drospirenone 3 mg) extended regimen (flexible) substantially alleviated both dysmenorrhea and also non-menstrual pain as well as deep dyspareunia in endometriosis patients. 10 Those findings were confirmed by data from a recent systematic-review. 11 Other researches have indicated that use of OCs for longer than 6 months can significantly reduce the size of ovarian endometriomas. 6,10, 12 Seracchioli et al found that both cyclic and continuous OC use effectively reduces the size of endometrioma at the time of recurrence and delays endometrioma recurrence. 13 Other research suggested that instead of producing cytoreductive effects, hormonal agents simply suppress cell proliferation and induce a cytostatic state in the ectopic endometrium as long as they are continued. 5 OC use, particularly extended or continuous use, can control not only the pain associated with endometriosis but also the progression of lesion in various endometriosis subtypes.
Despite potential advantages, however, cyclic and extended OC regimens carry the risk of side effects that can lead to decreased compliance or to serious deep vein thrombosis. Cyclic regimens are frequently associated with hormone withdrawal symptoms such as headache, bloating, nausea, and breast tenderness during the hormone-free intervals. 14 Extended and continuous OC regimens have been associated with a high-risk of breakthrough bleeding. 14 In addition, patients may refuse to continue a regimen due to anxiety associated with the loss of regular periods. 15 However, some women find the absence of menstrual blood flow and associated relief from unwanted symptoms beneficial.
In the present study, we conducted a placebo-controlled RCT to evaluate the efficacy and safety of a 28-day Cyclic regimen and 84day Extended regimen of NPC-16 in patients with dysmenorrhea.
The primary end point was the efficacy of the Cyclic and Extended NPC-16 regimens compared to the Placebo regimen. As a secondary end point, we compared the efficacy of the Cyclic and Extended regimens.

| Study design
This multicenter, randomized, placebo-controlled trial was conducted between February 2015 and January 2017. Patients with dysmenorrhea were enrolled from 18 private clinics across Japan.
Treatment started within the first 5 days of each patient's menstrual cycle. Patients allocated to the NPC-16-Cyclic regimen (Cyclic group) received NPC-16 for 21 days, followed by the placebo for 7 days (one cycle: 28 days); this schedule was repeated for 13 cycles. Patients allocated to the NPC-16-Extended regimen (Extended group) received NPC-16 for 77 days, followed by the placebo for 7 days (ie, one 84-day cycle, equivalent to three 28-day cycles), and this schedule was repeated for four cycles, followed by one cycle of the Cyclic regimen. After completion of the Extended regimen, an additional cycle of the Cyclic regimen was performed, because safety data for 1-year (13-cycle) treatment were required based on a mandate from the authorities. Safety test for drugs was requested to be continued for one year. Patients allocated to the Placebo group received the placebo for 28 days (one cycle); this schedule was repeated for four cycles, followed by nine cycles of the Cyclic regimen ( Figure 1). The placebo drugs were administered only four cycles due to the ethical reason for the placebo group.
NPC-16 (Cyclic and Extended regimens) and the placebo were prepared by the manufacturer (Nobelpharma Co., Ltd.) and supplied in 28-day blister packs of identical appearance. Use of non-hormonal agents for analgesic purposes was allowed, at the patient's discretion.
However, the use of hormonal drugs other than the trial drug was prohibited. Throughout the term of the study, the use of reliable contraception other than hormonal agents was required for all patients.
Vital signs (blood pressure and body weight) were measured for each patient in each cycle. Clinical laboratory tests, including hematology, blood biochemistry, coagulation and urinalysis, and transvaginal ultrasonography were carried out before treatment and at cycles 3, 6, 9, and 13. Uterine bleeding was evaluated based on entries in diaries recorded by patients throughout the study period. The degree of bleeding was assessed on a 5-point scale as follows: 0: none, 1: minimal (equivalent to spotting), 2: milder than usual menstruation, 3: similar to usual menstruation; and 4: heavier than usual menstruation.

| Study population
Of 313 patients screened, 62 were excluded before randomization, and the remaining 251 patients were randomly assigned.
Patients were enrolled if they met the following inclusion criteria:

| Determination of sample size
The study was designed to simultaneously compare differences in efficacy between the NPC-16-Cyclic and Extended groups versus the Placebo group. Because the Extended regimen was expected to produce a greater change in total dysmenorrhea score compared with the Cyclic regimen, we assumed that a comparison of the Cyclic and

| Randomization and blinding
Patients were randomized to receive the Cyclic regimen, Extended regimen, or Placebo (ratio, 1:1:1). Randomization was stratified by primary and secondary dysmenorrhea (ratio, 1:1). Randomization was carried out according to the permuted block method by a company engaged by Nobelpharma Co., Ltd. For patients with either primary or secondary dysmenorrhea, one block (representing six allocations; two for the Cyclic regimen, two for the Extended regimen, and two for the Placebo) was prepared and allocated to each of the 18 study sites. Information regarding allocation was restricted to the company that carried out the randomization and only after all data had been collected. Both the patients and physicians were blinded to the group to which each patient had been allocated.

| Efficacy
Patients were requested to visit the hospital before, during, and after treatment (total of 19 visits). Baseline data were collected at the pretreatment visit.
The primary end point was total dysmenorrhea score, calculated as the sum of separate scores for limitation of ability to work or study (pain score) and analgesic requirement (drug score). Scores were obtained using verbal rating scales developed by Harada  NPC-16-Cyclic Drug N P N P N P N P N P N P N P N P N P N P N P N P N P Days 2 1 7 21 7 21 7 21 7 21 7 21 7 21 7 21 7 21 7 21 7 21 7 21 7 21 1 7 21 7 21 7 21 7 21 7 21 7 21 7 21 7 21  For the pain score, patients were asked to assess their menstrual pain based on the number of days it limited their ability to work or study according to the following four-point rating scales: 0, none; 1, mild pain (some loss of work [or study] efficiency); 2, moderate pain (some need to rest in bed, loss of ability to work [or study]); and 3, severe pain (in bed for more than 1 day).
For the drug score, patients were asked to state the number of days they required analgesic drugs to relieve their pain, using similar four-point rating scale: 0, none; 1, 1 day; 2, 2 days; and 3, ≥3 days.
The secondary end point was the degree of dysmenorrhea as evaluated by patients using a visual-analog scale (VAS; range, 0-100).
In cases of absence of menstruation, the total dysmenorrhea score and VAS score were recorded as 0. Pain without menstruation was recorded as pelvic pain.
To evaluate the primary end point, the mean difference in total dysmenorrhea score between the mean score at baseline and the mean score after treatment (mean of cycles 1-3) was calculated, and comparisons were made between the Cyclic group or the Extended group and the Placebo group.
The change in VAS score from baseline to cycle 3 (mean of cycles

1-3) was compared between the Cyclic or Extended groups and the
Placebo group as the secondary end point.

| Adverse effects
Adverse events (AEs) were defined as any unfavorable or unintended clinical signs (including abnormal laboratory values), symptoms or diseases. ADRs were defined as AEs considered related to administration of the study drug.

| Statistical methods
Efficacy was evaluated using the full analysis set.

| Efficacy
One patient from the Cyclic group and two patients from the Placebo group were excluded from the efficacy analysis population because their efficacy data were not collected. Table 2 shows the change from baseline (difference between baseline and the mean of cycles 1-3) in the total dysmenorrhea and VAS scores.

| Total dysmenorrhea score
Overall, the reduction in total dysmenorrhea score was significantly greater in both the Cyclic group (-1.8) and Extended group (-3.1) than in the Placebo group (-0.9; P<.01) among patients with dysmenorrhea ( Table 2).
As shown in Figure 2A, the reduction in total dysmenorrhea score was greater in the Extended group than Cyclic and Placebo group. Changes in total dysmenorrhea score in the Cyclic and Extended groups according to severity of dysmenorrhea at baseline as shown in Figure 2C,D. In patients with moderate dysmenorrhea (total dysmenorrhea score, 3 or 4), the reduction in total dysmenorrhea score was greater in the Extended group (−2.6) than the Cyclic group (−1.8). In patients with severe dysmenorrhea (total dysmenorrhea score, 5 or 6), the reduction in total dysmenorrhea score was more notable in the Extended group (−3.8) than in the Cyclic group (−1.9; Table 2).
We also directly compared changes in dysmenorrhea score from baseline between the Cyclic and Extended groups. As clearly illustrated in Figure 3, the dysmenorrhea scores for each respective 3-cycle interval were significantly reduced in the Extended group

| VAS score
Overall, the reduction in VAS score among patients with dysmenorrhea was significantly greater in the Cyclic group (-22.6) and Extended group (-39.7) than the Placebo group (-9.5; P<.01; Table 2).
As shown in Figure 2B, the reduction in VAS score was greater in the Extended group than the Cyclic group. The VAS score for nonmenstrual pelvic pain was significantly reduced from baseline in both the Cyclic and Extended groups (data not shown).

| D ISCUSS I ON
We conducted the present RCT to evaluate the efficacy of Cyclic and Extended regimens of NPC-16 (containing EE plus levonorgestrel) compared with Placebo in patients with dysmenorrhea.
TA B L E 2 Efficacy of treatment: change from baseline (difference between baseline and the mean of cycles 1-3)

NPC-16-Cyclic NPC-16-Extended Placebo
Total dysmenorrhea score The present RCT also compared the efficacy of the Cyclic and Extended regimens with respect to the severity of dysmenorrhea. In patients with severe dysmenorrhea at baseline (total dysmenorrhea score, 5 or 6), the Extended regimen produced a more pronounced reduction in the total dysmenorrhea score compared to the Cyclic regimen. Although high-quality evidence of the effects of OCs on dysmenorrhea is limited, the results of the present study provide further support for the hypothesis that the Extended regimen is superior in the treatment of patients with severe dysmenorrhea involving endometriosis.
Primary and secondary dysmenorrhea patients were treated together in the present study. In modern medical management of endometriosis, hormone treatment is commenced when patients see physicians before a surgical diagnosis of endometriosis is made. 4 Thus, dysmenorrhea regardless of whether it is primary or secondary is an indication for OC treatment in patients with suspected endometriosis.
Extended or continuous OC regimens were introduced for contraceptive purposes, because the 7-day hormone-free interval in cyclic regimens is associated with hormone withdrawal symptoms, including headache and mood swing. 14 Regular menstruation, however, is also associated with dysmenorrhea and other related symptoms. suggested that progestins are preferable to OCs as a first-line treatment, the efficacy of OCs may be underestimated in the literature due to a lack of data regarding non-menstrual pelvic pain and deep pain, including dyspareunia and dyschezia. 21,22 Previously and in the present study, we demonstrated that both cyclic and extended OC regimens improve dysmenorrhea and also relieve non-menstrual pelvic pain and dyspareunia when used long-term. 10 Additionally, when progestin is used for many years, hypoestrogenic symptoms, including a reduction in bone mineral density, can be of concern, especially in adolescent and women in their 20s or 30s. 23 EE decreases the likelihood of irregular uterine bleeding, an ADR associated with progestin. 25 However, EE also increases the risk of venous thromboembolism, which although rare is potentially fatal.
In the present study, the incidence of irregular uterine bleeding was higher in the Extended group than in the Cyclic group but tended to decrease over time.
Continuous and extended OC regimens are recommended for treating pain in endometriosis patients, but high-quality background evidence of efficacy is still lacking. The efficacy of low-dose OC Extended regimen was compared directly with that of the Cyclic regimen in patients with secondary dysmenorrhea, including that caused by endometriosis. The results showed for the first time that the dysmenorrhea scores significantly decreased in the Extended group compared to the Cyclic group. Moreover, the new finding in the present study is that the NPC-16-Extended regimen may be more beneficial for patients with severe dysmenorrhea. But it was evaluated as a secondary end point in the present study. This is a weak point of the present study.
The present study confirmed that compared with placebo, administration of the low-dose OC NPC-16 as a Cyclic or Extended regimen, is efficacious for treating dysmenorrhea. The Extended regimen was more efficacious in the treatment of severe cases, although it was associated with a higher incidence of irregular uterine bleeding. NPC-16 is a useful option for the treatment of both primary and secondary dysmenorrhea, including that associated with endometriosis.

ACK N OWLED G EM ENTS
The authors thank the following physicians for participation in the

Conflicts of interest: Tasuku Harada and Mikio Momoeda received
consulting fees from Nobelpharma Co., Ltd.

E TH I C A L A PPROVA L
The study protocol (number NPC-16-2) was approved by the institutional review board at each study site. Throughout the course of the trial, monitors made regular visits to each study site to ensure adherence to the protocol. All patients provided written informed consent.