Efficacy and Safety of 48-Week Low-Dose Dienogest Treatment in Patients with Endometriosis-Associated Dysmenorrhea: A Randomized, Open-Label, Parallel-Group Trial

Why carry out this study? Endometriosis-associated dysmenorrhea imposes a significant burden on quality of life, and long-term treatment options with favorable safety profiles are needed. Both 1-mg/day and 2-mg/day doses of dienogest (DNG) are used in clinical practice in Japan; however, lower doses may reduce the risk of adverse hypoestrogenic effects and improve long-term adherence. This randomized controlled trial evaluated the efficacy and safety of 1 mg/day versus 2 mg/day of DNG over a 48-week treatment period. What was learned from the study? Both doses significantly reduced menstrual pain, using a visual analog scale (VAS), but the 1-mg group did not meet the predefined margin for non-inferiority to the 2-mg group. The findings support dose selection based on the balance between efficacy and safety and highlight the need for larger studies to confirm long-term safety (particularly bone mineral density reservation) and to evaluate efficacy in broader populations, including non-ovarian endometriosis.

This randomized, open-label, parallel-group, dose-comparison, multicenter trial was conducted at six centers in Japan (trial registration number: jRCTs041210016). Ethics: The study protocol was approved by the Clinical Research Review Committee of the Nagoya City University Hospital (approval number: 2021A001-23b001), and informed consent was obtained from all participants before enrollment in the study. This study adhered to the ethical principles of the Declaration of Helsinki. To ensure that the study participants had endometriosis, eligibility required the presence of at least one ovarian endometriotic cyst confirmed by imaging. Other phenotypes of endometriosis, such as deep infiltrating or peritoneal lesions, were not systematically assessed and therefore may have been present in the study population. Patients who met the inclusion and exclusion criteria were invited to participate in the study by the site investigators (Table  1 ). Table 1 Inclusion and exclusion criteria Inclusion criteria (patients meeting all of these criteria were included in this study) (1) Patients who have been diagnosed as having endometriosis (ovarian endometriotic cyst) and have not planned surgery (2) Patients with dysmenorrhea (3) Patients who plan to use DNG for 48 weeks (4) Patients with a regular menstrual cycle and aged 20 years or older at the time of informed consent (5) Patients who have provided written consent to participate in this study Exclusion criteria (patients meeting any of these criteria were excluded from this study) (1) Patients who have difficulty verbalizing subjective symptoms related to menstruation, etc (2) Patients currently under treatment for malignant diseases (3) Patients who have taken DNG, LEP preparations, or other hormonal drugs within 4 weeks of informed consent (4) Patients in whom it has been difficult to diagnose endometriosis (5) Patients deemed inappropriate for the study by the study investigator DNG dienogest, LEP low-dose estrogen progestin Inclusion and exclusion criteria DNG dienogest, LEP low-dose estrogen progestin The study design, eligibility criteria, endpoints, and statistical analysis plan were prespecified in a protocol previously published in JMIR Research Protocols [ 14 ]. Full methodological details are available in that publication; key elements are summarized below. Study data were collected using a combination of patient questionnaires, laboratory tests, bone density measurements, and imaging studies, including pre-registration magnetic resonance imaging (MRI) and ultrasound examinations. Once eligibility was confirmed, study personnel entered the patient information into the electronic data capture system to facilitate enrollment and randomization. Study data was collected and managed using Research Electronic Data Capture tools hosted at Gifu University Hospital, a secure, web-based platform that supports validated data entry, audit trails, automated data export, and system interoperability [ 15 , 16 ]. Randomization using the permuted block method with a block size of two or four was stratified on the basis of allocation factors, e.g., patient age and the presence or absence of uterine adenomyosis. Treatment commenced on the second to fifth day of the first menstrual cycle following enrollment. Patients were randomly assigned to receive either 0.5 mg DNG (1-mg group) or 1 mg DNG (2-mg group), administered orally twice daily for up to 48 weeks. Follow-up visits were scheduled 4 weeks post-enrollment and every 12 weeks thereafter. At each follow-up visit, comprehensive assessments were conducted, including patient questionnaires, vital sign measurements, blood tests, and ultrasound examinations to evaluate the size of the ovarian endometriotic cysts. An MRI scan was performed before study enrollment to establish baseline imaging data. Pain and estrogen deficiency symptoms were assessed using standardized questionnaires. Dysmenorrhea-related pain was evaluated using the visual analog scale (VAS) and modified dysmenorrhea score based on the study by Harada et al. [ 17 , 18 ] (Table  2 ). Throughout the manuscript, “menstrual pain” denotes the VAS outcome, whereas “dysmenorrhea” denotes the composite score. Estrogen deficiency symptoms were assessed using the Kupperman menopausal index (Table  3 ). Table 2 Dysmenorrhea score Degree Definition Score Pain None No pain 0 Slight Some difficulty with work, household chores, and schoolwork 1 Moderate Interference with work, housework, or schoolwork to the point of needing to lie down and take a break 2 Severe Sleeping for more than 1 day and unable to work, do housework, or study 3 Painkiller usage None None 0 Slight Used painkillers for 1 day during the menstrual period 1 Moderate Used painkillers for 2 days during the menstrual period 2 Severe Used painkillers for 3 days during the menstrual period 3 The dysmenorrhea score (total score: 0–6) is the sum of the subscores for the severity of pain and the use of analgesics Table 3 Kupperman menopausal index Symptom Degree Evaluation Hot and sweaty face 3・2・1・0 4 Numbness in the limbs and loss of sensation 3・2・1・0 2 Difficulty falling asleep and waking easily 3・2・1・0 2 Easily agitated and nervous 3・2・1・0 2 Worried and depressed 3・2・1・0 2 Feeling dizzy or nauseous 3・2・1・0 1 Tire easily 3・2・1・0 1 Pain in the shoulders, knees, and knots in the hands and feet 3・2・1・0 1 Headache 3・2・1・0 1 Palpitations 3・2・1・0 1 Feels like ants crawling on the skin 3・2・1・0 1 To determine the score, multiply the Degree of Symptomatology (Degree) number by the Rating Scale number (Evaluation) Scores were classified as follows: mild, 16–20; moderate, 21–34; and severe, ≥ 35. Degree was defined as strong (3), moderate (2), weak (1), or none (0) Dysmenorrhea score The dysmenorrhea score (total score: 0–6) is the sum of the subscores for the severity of pain and the use of analgesics Kupperman menopausal index To determine the score, multiply the Degree of Symptomatology (Degree) number by the Rating Scale number (Evaluation) Scores were classified as follows: mild, 16–20; moderate, 21–34; and severe, ≥ 35. Degree was defined as strong (3), moderate (2), weak (1), or none (0) Patients were allowed to use analgesics as needed for the management of dysmenorrhea-related pain. Bone density was assessed twice, at the beginning and at the end of the study period, using dual-energy X-ray absorptiometry. The primary efficacy endpoint was the change in VAS score for menstrual pain from baseline to week 48, calculated as the baseline score minus week 48 score. Key secondary efficacy endpoints included the change in the dysmenorrhea score and the rate of reduction in the size of ovarian endometriotic cysts, as assessed by ultrasonography and/or MRI, from baseline to week 48. Additional efficacy assessments included changes in the VAS score, dysmenorrhea score, and cyst size at weeks 12, 24, and 36. Safety endpoints were recorded as the incidence of low estrogen-related symptoms (e.g., headache, fatigue, palpitations, musculoskeletal stiffness, depressed mood, dizziness, hot flashes, anxiety, and insomnia) that were not captured by the Kupperman menopausal index. Changes in the Kupperman menopausal index, bone mineral density (BMD) (g/cm 2 ), and serum estradiol (E2) levels from baseline to week 48 were also recorded. Changes in the Kuperman score, BMD, and serum E2 level were also assessed at weeks 12, 24, and 36. The primary analysis was performed using a per-protocol set (PPS) to account for the possibility that stricter protocol adherence would yield a more conservative estimate of the treatment effect. The last observation carried forward method was used for missing data. The primary analysis involved comparing the change in the VAS score from baseline to 48 weeks between the 1-mg/day and 2-mg/day groups using the mean difference and its 95% confidence interval (CI). When the lower limit of confidence exceeded – 15 (non-inferiority margin), the non-inferiority of the effect of the 1-mg group to that of the 2-mg group was recognized. For the secondary efficacy endpoint, the 1-mg and 2-mg groups were compared to estimate the difference and its 95% CI for the mean value of the dysmenorrhea score and change in endometriotic cyst size from baseline to 48 weeks. In the safety evaluation, the frequency of low estrogen symptoms during the study treatment period was aggregated and a 95% CI was calculated on the basis of a normal approximation of the binomial distribution. Additionally, the chi-square test was used to compare the groups. The two-sided significance level was set at 5%. The frequencies of the occurrence of adverse events and side effects were calculated for safety analysis of the target population. As a secondary safety analysis, each group was compared to estimate the difference and 95% CI for the mean values of BMD, Kupperman menopausal index, and serum E2 level. Statistical analyses were performed using R software (version 4.1.0; R Foundation for Statistical Computing, Vienna, Austria). Based on a Japanese study [ 13 ], the mean change in the VAS score after 52 weeks with 1-mg DNG was − 52.1 [standard deviation (SD): 24.13] for secondary dysmenorrhea. In the Japanese phase II dose-finding study [ 19 ], the average changes in the VAS score were − 35.74 at 0.25 mg × 2, − 44.31 at 0.5 mg × 2, and − 50.78 at 1 mg × 2 at 12 weeks. Therefore, it was assumed that at least an effect of 0.25 mg × 2 or more could be expected by this test administration, and the non-inferiority margin was set to 15. It was also assumed that the effects in the groups would be similar, and the SD was assumed to be 24.13 with reference to a Japanese phase III study [ 12 ]. In this setting, 41 patients in each group (82 patients in total) were required to achieve 80% power at a one-sided significance level of 2.5%. When 5% of the dropout cases were considered in each group, the required sample size became 44 cases in each group (88 cases in total). We evaluated the incidence of low estrogen symptoms during the study treatment period as an important secondary endpoint. On the basis of the long-term use of DNG [ 13 ], we assumed the incidence proportions of 1-mg/day and 2-mg/day DNG to be 20% and 50%, respectively, with the chi-square test providing more than 80% power at a two-sided significance level of 5%. Therefore, the sample size of this study (88 patients in total) was sufficient for assessing the incidence of low estrogen symptoms. The study protocol was approved by the Clinical Research Review Committee of the Nagoya City University Hospital (approval number: 2021A001-23b001), and informed consent was obtained from all participants before study enrollment. This study adhered to the ethical principles of the Declaration of Helsinki. All data were anonymized to protect the participants’ privacy. No personally identifiable information is included in the dataset. Additionally, stringent protective measures, e.g., encrypted data storage and restricted access protocols, were implemented to ensure data confidentiality. The protocol was finalized before enrollment began.

A total of 88 participants were enrolled and randomly assigned to receive either 1-mg/day ( n  = 42) or 2-mg/day ( n  = 46) DNG (Fig.  1 ). Among them, 87 participants received at least one dose of the study medication and were included in the safety analysis, whereas one participant withdrew before treatment initiation. The full analysis set and PPS were defined according to the protocol compliance criteria. The baseline demographic and clinical characteristics were comparable between the groups (Table  4 ). Fig. 1 Participant flow. Participants who received at least one dose and had evaluable visual analog scale data. Safety analysis: participants who received at least one dose of the study drug Table 4 Demographic and clinical characteristics of the participants Variable (unit), mean ± SD or median (IQR) n Overall ( N  = 79) 1 Dienogest, 1 mg ( n  = 37) Dienogest, 2 mg ( n  = 42) P -value 2 Age at informed consent (y) 79 40.4 ± 8.1 40.6 ± 7.7 40.2 ± 8.5 0.788 Body weight (kg) 76 59.6 ± 12.9 62.8 ± 13.6 56.5 ± 11.7 0.287 Body mass index (kg/m 2 ) 65 23.5 ± 5.3 24.9 ± 5.3 22.4 ± 5.0 0.059 Uterine transverse diameter (cm) 68 5.4 (4.6–6.7) 5.7 (4.7–6.9) 5.2 (4.3–6.6) 0.376 Uterine longitudinal diameter (cm) 68 7.5 (6.4–8.6) 7.1 (6.4–8.4) 7.7 (6.4–8.7) 1 Dysmenorrhea score, mean ± SD 79 3.08 ± 1.85 2.85 ± 1.83 3.31 ± 1.87 0.275 Dysmenorrhea score, median (IQR) 79 3 (2–5) 3 (2–5) 3 (2–4) 0.256 Baseline pain VAS (0–100 mm) 79 64.0 (30.0–80.0) 60.0 (30.0–79.0) 70.0 (39.0–80.0) 0.599 Serum CA-125 level (U mL⁻ 1 ) 79 29.7 (21.2–55.7) 32.3 (22.3–60.7) 28.1 (18.1– 9.5) 0.293 Hemoglobin level (g/dL) 79 12.6 (11.5–3.6) 12.5 (11.5–13.3) 12.7 (11.5–13.6) 0.626 Serum estradiol level (pg/mL) 79 115.0 (60.0–206.0) 113.0 (53.0–257.0) 133.0 (61.0–183.0) 0.988 ALT level (U/L) 79 14 (11–20) 14 (10–9) 13 (11–0) 0.813 AST level (U/L) 79 17(14 – 22) 17 (14–19) 17(15–2) 0.484 BMD (g/cm 2 ) 66 1.0 (0.9–1.1) 1.0 (0.9–1.1) 1.0 (0.9–1.2) 0.39 ALT alanine aminotransferase, AST aspartate aminotransferase, BMD bone mineral density, BMI body mass index, CA-125 cancer antigen 125, IQR interquartile range, SD standard deviation, VAS visual analog scale 1 n  number of participants with available data for each variable; denominators vary because of missing height or laboratory data 2 P values: Welch t test for mean ± SD, Wilcoxon rank-sum (Mann–Whitney U ) test for median (IQR), and Fisher's exact test for categorical variables; no adjustment for multiple comparisons 3 All participants started DNG on cycle day 1–5; therefore, baseline cycle-day data are not shown 4 No participant received hormonal therapy within 4 weeks before randomization (exclusion criterion) Participant flow. Participants who received at least one dose and had evaluable visual analog scale data. Safety analysis: participants who received at least one dose of the study drug Demographic and clinical characteristics of the participants ALT alanine aminotransferase, AST aspartate aminotransferase, BMD bone mineral density, BMI body mass index, CA-125 cancer antigen 125, IQR interquartile range, SD standard deviation, VAS visual analog scale 1 n  number of participants with available data for each variable; denominators vary because of missing height or laboratory data 2 P values: Welch t test for mean ± SD, Wilcoxon rank-sum (Mann–Whitney U ) test for median (IQR), and Fisher's exact test for categorical variables; no adjustment for multiple comparisons 3 All participants started DNG on cycle day 1–5; therefore, baseline cycle-day data are not shown 4 No participant received hormonal therapy within 4 weeks before randomization (exclusion criterion) The primary efficacy endpoint was the change in the VAS score for menstrual pain from baseline to week 48. Both the groups demonstrated significant improvements. In the 1-mg group, the mean change from baseline was 44.63 mm (95% CI: 35.17–54.09), whereas in the 2-mg group, the mean change was 54.19 mm (95% CI: 44.75–63.64). The between-group difference was – 9.57 mm (95% CI: − 22.7 to 3.56). Although the 1-mg group showed pain reduction comparable to that of the 2-mg group, the lower bound of the 95% CI (− 22.7 mm) for the between-group difference did not exceed the predefined non-inferiority margin of − 15 mm. Therefore, the non-inferiority of the 1-mg dose could not be statistically confirmed. At week 48, the mean changes in the dysmenorrhea score from baseline were − 2.09 (95% CI: − 2.67 to − 1.51) and − 2.44 (95% CI: − 3.06 to − 1.83) in the 1-mg and 2-mg groups, respectively. The between-group difference was 0.36 (95% CI: − 0.47 to 1.19), indicating no statistically significant difference in the reduction of dysmenorrhea scores between the groups. At week 48, the mean absolute reductions in ovarian endometrioma volume were − 0.37 cm 3 (95% CI: − 0.62 to − 0.12) and − 0.5 cm 3 (95% CI: − 0.81 to − 0.18) in the 1-mg and 2-mg groups, respectively. The between-group difference in the mean volume change was 0.13 cm 3 (95% CI: − 0.27 to 0.52), with no statistically significant difference observed. Regarding the percentage reduction in the cyst volume, the mean reduction rates were 42.1% ± 62.5% and 64.2% ± 42.6% in the 1-mg and 2-mg groups, respectively. The corresponding median reduction rates were 64.1% [interquartile range (IQR): 25.1–86.4%] and 78.6% (IQR: 51.2–88.6%), respectively. Serum E2 levels decreased substantially in both groups throughout the treatment period. At week 48, the mean reductions in serum E2 levels were − 90.45 pg/mL (95% CI: − 152.51 to − 28.39) and − 104.73 pg/mL (95% CI: − 150.60 to − 58.86) in the 1-mg and 2-mg groups, respectively. The between-group difference was 14.28 pg/mL (95% CI: − 61.07 to 89.63), indicating no statistically significant difference in E2 level suppression between the groups. At week 48, the mean changes in lumbar spine BMD were − 0.005 g/cm 2 and − 0.024 g/cm 2 in the 1-mg and 2-mg groups, respectively, with a between-group difference of + 0.019 g/cm 2 (95% CI: 0 to 0). Changes in the Kupperman menopausal indexes over the 48-week treatment period were comparable between the groups. At week 48, the mean changes from baseline were – 4.95 (95% CI: − 7.09 to − 2.82) and − 4.93 (95% CI: − 7.45 to − 2.42) in the 1-mg and 2-mg groups, respectively. The between-group difference was − 0.02 (95% CI: − 3.29 to 3.26), indicating no statistically significant difference. No serious adverse events related to estrogen deficiency were observed during the 48-week treatment period. The frequency of hypoestrogenic symptoms was low, and no participants discontinued treatment because of these symptoms. Table  5 presents a detailed breakdown of symptom incidence. Table 5 Adverse events observed during the 48-week treatment period Symptom Dienogest, 1 mg Dienogest, 2 mg n % n % Shoulder stiffness 18 45 18 46.2 Headache 14 35 16 41 Fatigue 6 15 5 12.8 Vertigo (floating) 5 12.5 4 10.3 Palpitations 5 12.5 4 10.3 Drowsiness 5 12.5 7 17.9 Anxiety 4 10 5 12.8 Hot flashes 3 7.5 4 10.3 Emotional instability 3 7.5 2 5.1 Depressed mood 2 5 3 7.7 Hyperhidrosis 2 5 2 5.1 Irritability 2 5 3 7.7 Depression 1 2.5 3 7.7 Insomnia 1 2.5 1 2.6 Mood swings 1 2.5 1 2.6 Tenosynovitis 1 2.5 0 0 Depressive symptoms 1 2.5 2 5.1 Joint stiffness 1 2.5 1 2.6 Vertigo (rotational) 0 0 2 5.1 Nausea 0 0 1 2.6 No serious adverse events related to estrogen deficiency were observed during the 48-week treatment period Adverse events observed during the 48-week treatment period No serious adverse events related to estrogen deficiency were observed during the 48-week treatment period Treatment completion rates were 95.2% (40/42) and 84.8% (39/46) in the 1-mg and 2-mg groups, respectively. In the 1-mg group, two participants (4.8%) discontinued the study owing to voluntary withdrawal of consent. In contrast, seven participants (15.2%) in the 2-mg group discontinued treatment: two owing to consent withdrawal (4.3%), one owing to ineligibility confirmed after enrollment (2.2%), two owing to adverse events (4.3%), one owing to non-attendance at follow-up (2.2%), and one owing to the discovery of a malignant disease post-enrollment that violated the exclusion criteria (2.2%). Thus, although the overall discontinuation rate was higher in the 2-mg group, only two discontinuations were due to adverse events, and the malignant disease was determined to be unrelated to study treatment. Three adverse events were reported in the 2-mg group: nausea at week 4 (resolved), genital bleeding at study completion (mild), and a case of malignant lymphoma, which was diagnosed after enrollment and led to discontinuation due to post-enrollment ineligibility because the patient no longer met the eligibility criteria. No adverse events were reported in the 1-mg group.

This randomized, open-label trial compared the efficacy and safety of DNG at 1 and 2 mg/day for 48 weeks in patients with endometriosis-associated dysmenorrhea. While both groups showed significant reductions in menstrual pain, the statistical non-inferiority of the 1-mg dose could not be confirmed for the primary endpoint. However, both treatment arms demonstrated clinically significant improvements in multiple outcomes. Previous studies have established the efficacy of 2-mg/day DNG in reducing endometriosis-related pain [ 6 – 9 , 19 ]. In our trial, the mean VAS score reduction at week 48 in the 2-mg group was consistent with reductions reported by Strowitzki et al. [ 9 ] (47.5 mm) and Osuga et al. [ 19 ] (50.78 mm) in prior phase III trials. The observed dysmenorrhea score reduction in the 2-mg group also aligned well with earlier findings, reinforcing the robustness of our efficacy outcomes. Although limited, existing evidence on the lower 1-mg dose, particularly from studies on primary dysmenorrhea, has demonstrated significant symptom relief. In a trial by Osuga et al. [ 19 ], 1-mg/day DNG produced a VAS score reduction of – 44.31 mm after 12 weeks. Here, we observed a similar degree of pain relief over 48 weeks. Additionally, nearly half of the participants in the 1-mg group achieved complete pain resolution, a rate comparable to that in the 2-mg group. These findings support the clinical utility of lower doses, particularly in patients seeking long-term symptom control, with fewer adverse effects. Concerning lesion size, previous studies have indicated that 2-mg/day DNG reduces ovarian endometriotic cyst volume by 66–79% over 6–12 months [ 20 – 24 ]. Our results are consistent with these data. Importantly, the 1-mg group also achieved a meaningful reduction, despite the lower estrogen-suppressive effect, suggesting that half the standard dose may be sufficient to exert therapeutic effects on endometriotic lesions. A key finding was the difference in the onset of pain relief: the 2-mg group demonstrated a more rapid reduction in VAS scores in the early treatment phase than the 1-mg group. This may be clinically relevant for patients presenting with severe baseline symptoms or those requiring a faster response. In contrast, the 1-mg group showed the advantage of a milder hormonal suppression profile, which may be beneficial for long-term maintenance therapy. These observations support a personalized approach for dose selection on the basis of patient characteristics and treatment goals (Fig.  2 ). Fig. 2 Pain relief. The 1-mg dienogest (DNG) group is shown as a blue dotted line , and the 2-mg DNG group is represented as a red solid line . Mean changes from baseline in the visual analog scale (VAS) score are presented at each time point, with error bars indicating 95% confidence intervals. Time course of changes in the VAS score over 48 weeks in the 1-mg and 2-mg DNG groups. Data represent mean change from baseline with 95% confidence intervals. A greater reduction is observed in the 2-mg DNG group than in the 1-mg DNG group during the early treatment phase Pain relief. The 1-mg dienogest (DNG) group is shown as a blue dotted line , and the 2-mg DNG group is represented as a red solid line . Mean changes from baseline in the visual analog scale (VAS) score are presented at each time point, with error bars indicating 95% confidence intervals. Time course of changes in the VAS score over 48 weeks in the 1-mg and 2-mg DNG groups. Data represent mean change from baseline with 95% confidence intervals. A greater reduction is observed in the 2-mg DNG group than in the 1-mg DNG group during the early treatment phase From a safety perspective, both doses were well tolerated with no serious adverse events attributable to estrogen deficiency. The 2-mg group maintained serum E2 within the threshold range (30–50 pg/mL), consistent with Barbieri’s estrogen threshold theory [ 25 ]. Moreover, the 1-mg group maintained higher median E2 levels (approximately 63 pg/mL), likely contributing to the better preservation of BMD. Our data confirmed this hypothesis: BMD loss was – 2.85% in the 2-mg group but only − 0.46% in the 1-mg group, suggesting that, compared with 2-mg/day DNG, 1-mg/day DNG may be a safer long-term option for patients at risk of bone loss. Because this was a multicenter study with non-uniform DXA systems, BMD z-scores were not consistently available and were therefore not analyzed, which is a study limitation. Irregular bleeding, a known side effect of progestins, was reported in both groups, with a slightly higher frequency in the 1-mg group. However, most events were mild and self-limiting, and no patients required discontinuation. Additionally, Kupperman menopausal indexes improved in both groups, possibly reflecting the positive impact of pain control on patient-reported menopausal-like symptoms, e.g., fatigue and irritability. No clinically significant hypoestrogenic adverse events were observed during the study period. Although statistical non-inferiority of the 1-mg dose was not established, this study demonstrated that 1-mg/day DNG provided substantial pain relief, reduced lesion size, preserved bone density, and had a favorable safety profile. These attributes make it a viable option for patients who prioritize tolerability and long-term safety. Conversely, 2-mg/day DNG remains the preferred choice when rapid, more robust symptom control is required. Dose tailoring on the basis of clinical context and patient preferences is warranted. This study has some limitations. First, although the mean between-group difference in the VAS score was small, it did not meet the statistical criteria for non-inferiority, which may have affected the interpretation of comparative efficacy. Second, the open-label design introduces potential bias in patient-reported outcomes, although objective endpoints, e.g., BMD and imaging data, provide supportive evidence. Third, our findings apply specifically to patients with ovarian endometriosis and may not be generalizable to other phenotypes, e.g., deep infiltrating or peritoneal lesions. Fourth, the 48-week duration may be insufficient to capture longer-term skeletal effects; follow-up of at least 2 years would better characterize bone outcomes. Fifth, because this was a multicenter trial with non-uniform DXA systems, only absolute BMD values (g/cm 2 ) were consistently available and z-scores could not be analyzed.

Although 1-mg/day DNG demonstrated slightly lower efficacy in pain reduction than at 2-mg/day DNG, it was associated with a favorable safety profile and clinically meaningful improvements in endometriosis-related symptoms. These findings suggest that 1-mg/day DNG may be a viable long-term treatment option for patients in whom safety considerations are paramount, such as those at risk of adverse hypoestrogenic effects. Conversely, 2-mg/day DNG may be preferred when greater analgesic efficacy is required. This study provides important evidence supporting dose-selection strategies and may inform future considerations for expanding the therapeutic indications of low-dose DNG for the management of endometriosis. However, to further support clinical decision making, larger studies with increased statistical power may help clarify whether 1-mg/day DNG is truly non-inferior to 2-mg/day DNG or whether 2-mg/day DNG offers significantly superior pain relief.

Endometriosis is a chronic, estrogen-dependent gynecological disorder that impairs women’s quality of life and often requires long-term hormonal treatment to prevent recurrence [ 1 – 5 ]. Dienogest (DNG) 2 mg/day is widely used and has demonstrated efficacy in controlling pain and reducing lesion size [ 6 – 9 ]. DNG is a selective progestin with a low risk of thromboembolic events and is considered safe for use in women aged 40 years or older, a population generally considered to be at increased risk of thromboembolism [ 10 , 11 ]. In Japan, a 1-mg/day dose has been approved for primary dysmenorrhea, and previous studies have shown that this lower dose may suppress symptoms with milder estrogen reduction than a higher dose [ 12 , 13 ]. However, no prior trials have directly compared the efficacy and safety of 1-mg/day versus 2-mg/day DNG for endometriosis-associated dysmenorrhea over an extended treatment period. The rationale for evaluating low-dose DNG has been previously described in detail [ 14 ]. Therefore, this study aimed to compare the efficacy and safety of 1- mg/day and 2-mg/day DNG over a 48-week treatment period in patients with endometriosis-associated dysmenorrhea. Additionally, this study aimed to determine whether low-dose DNG could serve as a safe and effective long-term therapeutic option in clinical practice.

Below is the link to the electronic supplementary material. Supplementary file1 (PDF 212 KB) Supplementary file1 (PDF 212 KB)