The effectiveness of desogestrel for endometrial protection in women with abnormal uterine bleeding-ovulatory dysfunction: a non-inferiority randomized controlled trial

Participants. Figure 1 shows flow of study participants. Of 109 eligible patients, 104 cases were enrolled and randomized to the study group (DSG, n = 50) or the control group (MPA, n = 54). According to the exclusion criteria, one case was excluded, and four cases declined to participate. There were two dropouts in the MPA group; one case after treatment cycle-2 because of colon cancer, and the other after cycle-5 because of amenor- rhea. Their missing data regarding withdrawal bleeding were imputed from the last observation carried forward. Therefore, there were five imputed data, including four cycles of normal withdrawal bleeding and one cycle of no withdrawal bleeding; the data of which contributed to the clinical outcome of the MPA group. Table 1 shows demographic and baseline characteristics of the 104 intention-to-treat participants. The DSG and MPA groups were comparable in age (44.8 ± 5.7 vs. 42.5 ± 7.1 years), body mass index (BMI, 24.8 ± 4.7 vs. 24.9 ± 4.7 kg/m2), education > 12 years (64.0% vs. 61.1%), and AUB characteristics (irregular bleeding of 100%). Both groups were comparable in baseline metabolic parameters, including fasting blood glucose, total cholesterol, triglycerides, high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL-C) levels. The non-compliance rate was low in both groups (2.8% in the DSG and 2.6% in the MPA groups). Efficacy outcomes. Figure 2 and Table 2 show efficacy outcomes. Figure 2 shows various degrees of endo- metrial response to treatment. The treatment is considered effective if the post-treatment endometrium shows a small inactive gland with pseudodecidual stromal change, as in Fig.  2a. The endometrial response to DSG seemed to be slightly better than that to MPA as determined by the proportion of complete pseudodecidualiza- Assessed for eligibility( n = 109) Allocated to DSG group (n = 50): − Received DSG (n = 50) −D id not receive DSG (n = 0) Allocated to MPA group (n = 54): −R eceived MPA( n = 54) −D id not receive MPA (n = 0) Excluded (n = 5) − Not meet exclusion criteria (n = 1) −D eclined to participate (n = 4) Randomized (n = 104) Lost to follow-up (n = 0) Discontinued DSG (n = 0) Lost to follow-up (n = 0) Discontinued MPA( n = 2): −C olon cancer (n = 1, finished 2nd cycle) −A menorrhea (n = 1, finished 5th cycle) Analyzed (n = 50) −E ndometrial response (n = 50 cases) −C linical response (n = 250 cycles) Excluded from analysis (n = 0) Analyzed (n=54) −E ndometrial response (n = 54 cases) −C linical response (n = 265 cycles) Excluded from analysis (n = 0) Figure 1. CONSORT participant flow. DSG desogestrel, MPA medroxyprogesterone acetate. 3 Vol.:(0123456789)Scientific Reports | (2022) 12:1662 | https://doi.org/10.1038/s41598-022-05578-0 www.nature.com/scientificreports/ tion (78.0% vs. 70.4%, P = 0.375). The 95% CI of difference (Δ) of this outcome (− 9.1%, 24.4%) indicated the non-inferiority of DSG to MPA as its lower bound lay above the non-inferiority limit of − 10%. Both DSG and MPA effectively normalized bleeding to an acceptable level (normal or no withdrawal bleed- ing) for > 90% of cases. However, the clinical response to DSG was not as good as that to MPA, as the propor - tion of acceptable bleeding was significantly lower in the DSG group than in the MPA group (90.0% vs. 96.6%, P = 0.016). In other words, the DSG group had a significantly higher rate of prolonged bleeding (10.0% vs. 3.4%). Safety and tolerability. Table 3 shows safety and tolerability data which presents the incidence of adverse events and side effects, and the changes from baseline to month-6 of treatment (δ) for various metabolic param- eters. One patient in the MPA group was found to have colon cancer after treatment cycle-2 and was withdrawn Table 1. Demographic and baseline characteristics of 104 intention-to-treat patients. Data are mean ± standard deviation, median and interquartile range (in parentheses), or number and percentage (in parentheses), otherwise specified. BP blood pressure, DSG desogestrel, HDL-C high-density lipoprotein cholesterol, LDL-C low-density lipoprotein cholesterol; MPA, medroxyprogesterone acetate. Characteristics DSG (N = 50) MP A (N = 54) Age (years) 44.8 ± 5.7 42.5 ± 7.1 Body mass index (kg/m2) 24.8 ± 4.7 24.9 ± 4.7 Waist circumference (cm) 80.6 ± 9.5 79.6 ± 9.9 Parity, median (range) 2 (0 to 4) 1 (0 to 4) Systolic BP (mmHg) 121.7 ± 16.4 124.9 ± 17.0 Diastolic BP (mmHg) 72.4 ± 10.7 72.6 ± 10.9 Education > 12 years 32 (64.0) 33 (61.1) Medical diseases (at least one) 11 (22.0) 11 (20.4)  Autoimmune diseases 2 (4.0) 1 (1.9)  Hypertension 4 (8.0) 4 (7.4)  Thyroid diseases 5 (10.0) 3 (5.6)  Others 0 (0) 3 (5.6) Character of abnormal uterine bleeding  Irregular bleeding 36 (72.0) 47 (87.0)  Irregular and excessive bleeding 14 (28.0) 7 (13.0) Metabolic parameters  Fasting blood glucose (mg/dL) 89 (83 to 93) 85 (80 to 90)  Total cholesterol (mg/dL) 198 (178 to 220) 196 (171 to 216)  Triglycerides (mg/dL) 81 (60 to 106) 91 (67 to 143)  HDL-C (mg/dL) 63 (48 to 73) 65 (54 to 77)  LDL-C (mg/dL) 116 (103 to 137) 103 (86 to 129) (N = 250) (N = 265) Non-compliance treatment cycles  Skipped doses > 20% 7 (2.8) 7 (2.6) Figure 2. Histopathology of endometrial response after treatment. (a) Complete transformation showing a small inactive gland with pseudodecidual stromal change, (b) Proliferative endometrium without response, (c) Inadequate secretory change and inconspicuous pseudodecidual stromal change. 4 Vol:.(1234567890)Scientific Reports | (2022) 12:1662 | https://doi.org/10.1038/s41598-022-05578-0 www.nature.com/scientificreports/ from the trial. The most common side effect was drowsiness, which approximately one-third of patients reported in both groups. The incidence of nausea was clinically significantly higher in the DSG group (12.0% vs. 1.9%, P = 0.053). Both DSG and MPA groups were comparable in the δ of every metabolic parameter. Most of the δ had a minus sign indicating favorable changes for total cholesterol, triglycerides, and LDL-C, but an unfavorable change for HDL-C. None of the metabolic parameters had a δ beyond ± 10%; therefore, the δ was not clinically important.

Treatment for AUB-O aims to restore normal physiological change of the endometrium to prevent endome- trial neoplasia and normalize menstruation 2. There is a lack of evidence from randomized controlled trials, especially about the effect of treatment for AUB-O on endometrial histopathology. Most studies showed only intermediate outcomes, such as menstrual cycle regularization15–17. Our present study showed that two cyclic- progestin regimens (DSG 150 µg/d or MPA 10 mg/d, 10 d/month) could transform endometrium to complete pseudodecidualization in > 70% of cases in the treatment cycle-1 and could normalize menstruation in > 90% of all treatment cycles. Table 2. Efficacy outcomes. The endometrial response rates were compared using chi-square test; whereas the clinical response rates were compared using generalized estimating equations (GEE). A P-value < 0.05 indicates statistical significance. For endometrial response, 95% CI of difference (Δ) in the proportion of complete pseudodecidualization is (− 9.1% to 24.4%); the lower bound of which lies above the non-inferiority limit of − 10% For clinical response, acceptable bleeding includes treatment cycles with normal withdrawal bleeding (amount and duration) and treatment cycles without withdrawal bleeding. DSG desogestrel, MPA medroxyprogesterone acetate. Outcomes DSG M PA PN n (%) N n (%) Endometrial response (cases) 50 54 0.375  Complete pseudodecidualization 39 (78.0) 38 (70.4)  Partial or no pseudodecidualization 11 (22.0) 16 (29.6) Clinical response (cycles) 250 265 0.016  Acceptable bleeding 225 (90.0) 256 (96.6)  Prolonged bleeding 25 (10.0) 9 (3.4) Table 3. Safety and tolerability. The incidence of serious adverse events and side effects are presented in number and percentage (in parentheses). The changes from baseline to month-6 of treatment (δ) of metabolic parameters are presented in median and interquartile range (in parentheses). Categorical data were compared using chi-square test or Fisher’s exact test. Continuous data were compared using Mann–Whitney U test. The serious adverse event included one case of colon cancer in the MPA group. Dermatologic problems included acne, pigmentation, or oily skin. DSG desogestrel, MPA medroxyprogesterone acetate, HDL-C high-density lipoprotein cholesterol, LDL-C low-density lipoprotein cholesterol. N DSG N M PA P Serious adverse events 50 0 (0) 54 1 (1.9) 1.000 Side effect (at least one) 50 29 (58.0) 54 29 (53.7) 0.659  Breast pain 50 6 (12.0) 54 3 (5.6) 0.307  Dermatologic problems 50 11 (22.0) 54 10 (18.5) 0.659  Drowsiness 50 14 (28) 54 13 (24) 0.648  Fever-like symptom 50 0 (0) 54 1 (1.9) 1.000  Increased appetite 50 5 (10.0) 54 9 (16.7) 0.320  Nausea 50 6 (12.0) 54 1 (1.9) 0.053  Pelvic cramp 50 0 (0) 54 2 (3.7) 0.496  Rash 50 1 (2.0) 54 2 (3.7) 1.000 δ of metabolic parameters  Fasting blood glucose (mg/dL) 50 0 (− 4.3 to 4.3) 52 0.5 (− 2.8 to 4.8) 0.840  Total cholesterol (mg/dL) 50 − 8.5 (− 25.5 to 3.3) 52 − 0.5 (− 12.0 to 12.0) 0.072  Triglycerides (mg/dL) 50 − 7.5 (− 21.3 to 9.5) 52 − 6 (− 24.0 to 14.5) 0.723  HDL-C (mg/dL) 50 − 5 (− 10.3 to 0.3) 52 − 3.5 (− 12.0 to 1.8) 0.776  LDL-C (mg/dL) 50 − 1.7 (− 20.6 to 8.4) 51 5.6 (− 10 to 18.4) 0.159 5 Vol.:(0123456789)Scientific Reports | (2022) 12:1662 | https://doi.org/10.1038/s41598-022-05578-0 www.nature.com/scientificreports/ Obese women with AUB-O are at high risk for developing endometrial neoplasia and may benefit from pre- ventive measures18. Although authorities recommend that endometrial biopsy should be performed in women older than 45 years19,20, obese women may need endometrial biopsy at a younger age18. Our study population was slightly younger than 45 years, but they had BMI in the overweight to obese classification according to the WHO Asia Pacific BMI cut point (≥ 23 kg/M2) 21. The preventive measures for these women include bodyweight reduction by lifestyle modification and progestin therapy18. Progestin therapy is available in various forms, including combined hormonal contraceptives, progestin- only contraceptives, and cyclic-progestin therapy. Each form has its own benefits and drawbacks. The progestin therapy results in pseudodecidualization of the endometrium. A case–control study reported that progestins in most combined oral contraceptives were adequate to protect against endometrial cancer22. DSG in the form of progestin-only oral contraceptive might also have this protective effect as it contains a similar dosage of DSG as a combined oral contraceptive does. It is unknown whether cyclic progestin therapy in premenopausal women with AUB-O also retains this protective effect. Evidence from menopausal hormone therapy shows that continuous combined estrogen-progestin regimens reduce the risk of endometrial cancer while cyclic regimens do not alter the risk23. We presume that the risk of endometrial cancer in women with AUB-O treated with cyclic progestin may be the same as the risk in women with regular ovulation. The equivalent dosages of progestins that are adequate for endometrial transformation were reviewed by Schindler, et al. 11 The authors summarized the daily and total doses of progestins. They considered MPA 5–10 mg/d for a total dose of 80 mg and DSG 150 µg/d for a total dose of 2 mg to be equivalent. In the present study, a total dose of 100 mg MPA or 1.5 mg DSG was used within 10 days. Although the total dose of 1.5 mg DSG was less than that suggested by Schindler’s review, this dosage was as effective as 100 mg MPA. The 10-day application of one or the other progestins could not completely transform the endometrium in some patients. Ineffective endometrial transformation may be due to the inadequate cumulative effect of progestin: dosage and duration2. It is suggested that the duration of cyclic progestin therapy should be 12–14 days per cycle to protect the endometrium effectively2,12. However, this recommendation was based on information from the first-generation progestins. Therapy with more potent third-generation progestins may protect the endometrium with a shorter duration of treatment. We found that 10-day DSG seemed to be more effective than 10-day MPA (78% vs. 70% complete pseudodecidualization). However, our study did not have enough power to show the superiority of DSG over MPA in this regard. Despite the observed comparable effectiveness relating to endometrial response, MPA was superior to DSG for clinical response. The drawback of DSG was the higher rate of bleeding for longer than eight days (11.1% vs. 4.6%). Some participants might have concomitant structural lesions as a confounding cause for prolonged withdrawal bleeding. In the present study, transvaginal ultrasonography could exclude gross structural lesions but not a tiny endometrial polyp which might be detected by saline infusion sonohysterography or hysteroscopy. With the strength of RCT, this confounder should be distributed equally between the groups. Interestingly, the finding that MPA had better bleeding control than DSG was consistent with our previous study that found MPA superior to other progestins regarding menstruation normalization in women with non-atypical endometrial hyperplasia24. Since each progestin has its own unique profiles10,25,26, each may act differently on molecular microenvironment involving hemostasis of endometrium, which may cause different uterine bleeding patterns. Some progestins have an adverse effect on metabolic profiles, although most of the evidence is derived from studies of menopausal hormone therapy. A review in 2017 regarding progestins and blood lipids found that MPA mitigated the beneficial effects of menopausal hormone therapy on blood lipids27. In our study, we expected the metabolic profile to be better in the DSG group than in the MPA group because DSG has a low androgenic effect and little to no glucocorticoid or mineralocorticoid effect. However, the change in all metabolic parameters from baseline to month-6 was marginal and may not be clinically meaningful. It was possible that cyclical application of these two progestins for 10 d/month did not have an adverse effect on metabolic parameters. The strength of the present study is its prospective randomized controlled design that can prevent bias. Moreover, we evaluated both endometrial and clinical responses to ensure that the treatment not only normal- ized menstruation but also protected the endometrium. However, RCT has a limitation in generalizability. We evaluated two cyclic-progestin regimens in overweight premenopausal women with AUB-O and followed up them for 6 months. Application of these regimens to patients with different characteristics in a longer duration of treatment may result in different efficacy and safety outcomes. In conclusion, according to our definition of non-inferiority, the cyclical administration of DSG (150 µg/d, 10 d/month) is not inferior to MPA (10 mg/d, 10 d/month) for pseudodecidualization of endometrium in women with AUB-O. Both DSG and MPA are effective for menstruation normalization in > 90% of cases, but DSG is less effective than MPA. The 6-month metabolic profile changes are comparable between the DSG and MPA groups, and are not clinically meaningful. Therefore, cyclic DSG can be used as an alternative to MPA for endometrial protection and menstruation normalization in AUB-O.

This non-inferiority double-blind RCT was conducted at the Gynecologic Endocrinology Unit, Department of Obstetrics and Gynecology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand, from 11 March 2014 to 30 November 2018. This trial was a part of the Siriraj Abnormal Uterine Bleeding (SI-AUB) project, an ongoing project established in 2010. The SI-AUB project comprises many sub-studies, including observational studies and clinical trials. The SI-AUB project and this study were conducted in accordance with the principles set forth in the Declaration of Helsinki. 6 Vol:.(1234567890)Scientific Reports | (2022) 12:1662 | https://doi.org/10.1038/s41598-022-05578-0 www.nature.com/scientificreports/ Participants. Participants were premenopausal women who were newly diagnosed with AUB-O. A diag- nosis of AUB-O was made if a patient had unremarkable findings from pelvic ultrasonogram and endome- trial histopathology revealed no structural lesions28. Regarding the endometrial histopathology, only patients with proliferative endometrium were recruited. A participant was excluded if she had conditions that might cause AUB (e.g., having coagulation defect or taking certain medications), took hormonal medication within 3 months before enrollment, planned to get pregnant within 6 months, needed hormonal contraception, had contraindications for progestogen administration (e.g., history of breast cancer, severe liver disease, or progesto- gen allergy), or had diabetes mellitus and/or dyslipidemia with ongoing treatment. All sexually active partici- pants were advised to use barrier contraception. An enrolled participant would be withdrawn from the study if she had a follow-up endometrial histopathology report of endometrial neoplasia or she encountered a serious adverse event. This participant would be counted as a case of treatment failure in the intention-to-treat analysis. Study procedures. Eligible participants were recruited from the gynecology outpatient department. Rou- tine evaluation for AUB at our clinic includes clinical evaluation and pelvic ultrasonography. Endometrial histo- pathology is indicated for women at increased risk of endometrial neoplasia, e.g., older than 45 years, irregular menstruation in women with polycystic ovary syndrome, or obesity. Demographic characteristics, medical his- tory, menstrual bleeding pattern, concomitant medications, and findings of general and gynecologic examina- tions were collected using a structured record form. Transvaginal ultrasonography was performed to assess the structure of pelvic organs. When the ultrasound probe could not be inserted vaginally, transrectal ultrasonog- raphy would be performed instead. Endometrial tissue was obtained by an office endometrial biopsy procedure using a flexible instrument. Patients were then scheduled two weeks after the procedure to be enrolled into the present study or another sub-study of the SI-AUB project. The enrolled participants were scheduled for five visits, including visit-1 at enrollment; visit-2 on day-8, day- 9, or day-10 of treatment cycle-1; visit-3 at four weeks; visit-4 at 3 months; and visit-5 at 6 months. Visit-1 was for collecting baseline data, obtaining biochemical blood tests, issuing a menstrual diary card, and providing the cycle-1 medication. Visit-2 was for obtaining endometrial sampling. Visit-3 was for evaluating compliance and side effects, and for providing cycle-2 and cycle-3 medication. Visit-4 was for providing cycle-4 to cycle-6 medication. Visit-5 was for follow-up biochemical blood tests. All participants were instructed to record data relating to the following in their menstrual diary cards: day of taking trial medication, possible side effects, and vaginal bleeding. One diary card was used to record data from one treatment cycle. The participants had to return the diary card and the leftover medication at the fol - lowing visit. Trial medication and treatment allocation. The trial medication was 10  mg MPA or 150  µg DSG administered orally once daily before bedtime for 10 consecutive days per month for 6 months. Each dose of both study medications was encapsulated in a white capsule that was identical in physical appearance. Ten cap- sules of the same medicines were pre-packed in a zip-locked opaque plastic bag labeled with a serial number according to the randomization code. Patients were allocated to either the control group (MPA) or the study group (DSG) by simple randomization using a random allocation software program29. The randomization codes were individually contained in sealed opaque envelopes sequentially numbered. An independent nurse who had no contact with study participants opened each envelope and dispensed the assigned package of trial medica- tion. The packaging of trial medications and the generation of randomization codes were produced by an inde- pendent pharmacist of the Department of Pharmacy. Therefore, the gynecologists and nurses who contacted the patients, the patients themselves, and the gynecologic pathologist were blinded to the group assignments. Instruction for a missed dose of trial medication included taking the missed dose as soon as the patients remembered or skipping the dose when it was almost the time of the next scheduled dose. In the latter case, the participants would complete the trial medication in > 10 days. The cycles with > 20% skipped doses were counted as non-compliance treatment cycles. Outcomes. The primary efficacy outcome was the “endometrial response”, which was evaluated from his- topathology of endometrial biopsy on medication day-8, day-9, or day-10 of treatment cycle-1. Endometrial biopsy was performed using a disposable, flexible cannula with its own internal suction piston (Endocell®, Wal- lach Surgical Devices, Trumbull, CT, USA). The endometrial tissue was immediately fixed in formalin and sent to the Department of Pathology for routine tissue processing with hematoxylin and eosin staining. The tissue section was examined under a light microscope by a gynecologic pathologist (M.W .). Endometrial histopathol- ogy was categorized into (i) complete pseudodecidualization if the entire endometrial biopsy specimen showed pseudodecidual transformation, (ii) incomplete pseudodecidualization if the specimen showed uneven pseudo- decidual transformation, (iii) treatment failure if the specimen showed proliferative endometrium or endome- trial neoplasia, and (iv) undetermined if the specimen was inadequate for evaluation. The success rate of endo- metrial response was the proportion of the cases with complete pseudodecidualization in the intention-to-treat cases. Dropout cases were counted as treatment failure cases. The secondary efficacy outcome was the “clinical response” , which was evaluated from characteristics of with- drawal bleeding of treatment cycle 2 to cycle 6. The withdrawal bleeding was defined as bleeding that occurred after completion of each course of trial medication and before starting a new one (i.e., between days 11 and 28 of the treatment cycle). The amount of bleeding was categorized according to the participant’s experience30 into (a) light bleeding or spotting (less than normal menstruation), (b) normal bleeding (like normal menstruation), and (c) heavy bleeding (more than normal menstruation). The bleeding pattern was categorized1 into (i) normal bleeding (bleeding lasted up to eight days), (ii) prolonged bleeding (bleeding lasted more than eight days), (iii) 7 Vol.:(0123456789)Scientific Reports | (2022) 12:1662 | https://doi.org/10.1038/s41598-022-05578-0 www.nature.com/scientificreports/ breakthrough bleeding (bleeding occurred on the day of taking medication), and (iv) no bleeding (no bleeding during the medication-free days. The success rate of clinical response was the proportion of the treatment cycles with acceptable bleeding [i.e., the cycles with bleeding characteristics (i-a), (i-b), or (iv)] in the intention-to-treat cycles. Missing data were imputed from the last observation carried forward. Safety of treatment was evaluated from the changes in biochemical blood tests from baseline to month-6, and from side effects that occurred at any time during the study period. Biochemical blood tests were performed at visit-1 and visit-5. A venous blood sample was drawn from an antecubital vein during the 08.00 to 10.00 time period after overnight fasting for 12 h. The blood sample was examined for fasting blood glucose and lipid profile (total cholesterol, triglycerides, HDL-C, and LDL-C). All biochemical assays were performed at the Department of Clinical Pathology (central laboratory certified by IS0 15,189). All assays were performed using automatic analyzers, and all assays had intra-assay and inter-assay coefficients of variation (CV) of < 5%. Sample size. The sample size was calculated from the primary outcome to determine the non-inferiority of DSG compared with MPA regarding the success rate of endometrial response. A free online sample size calcula- tor was used for comparing two proportions: 2-sample non-inferiority or superiority (HyLown Consulting LLC, GA, USA). Given that the non-inferiority margin was − 10% (δ = − 0.1), the success rate of endometrial response derived from our pilot study in the DSG (θ1) and the MPA (θ2) groups was 80% and 60%, respectively. The required sample size with the equal allocation (r = 1) to achieve 90% power and an α of 0.025 was 47 patients per group. To compensate for 5% dropouts, the final sample size was 50 patients per group. Statistical analysis. Data were analyzed using SPSS Statistics software version 18.0 (SPSS, Inc., IL, USA). Statistical analyses of the efficacy outcomes were based on the intention-to-treat populations, and those of the safety outcomes were based on the per-protocol population. Data are presented as mean ± standard deviation (SD), median and interquartile range, or number and percent, as appropriate. Shapiro–Wilk test and Q-Q plot were used to examine the normality of continuous data. Unpaired t-test or Mann–Whitney U test was used to compare continuous data. Chi-square test or Fisher’s exact test was used to compare categorical data. General- ized estimating equations model was used to test the difference in the rate of acceptable bleeding from all treat- ment cycles between DSG and MPA groups. All statistical tests to determine differences between groups were 2-sided, and a P-value of less than 0.05 was considered statistically significant. Non-inferiority was considered if the lower bound of the 95% confidence interval (CI) for the difference (Δ) in the favorable outcome lay above the non-inferiority limit. Therefore, DSG would be considered non-inferior to MPA if the lower bound for the difference in the endometrial response rate lay above − 10%. Ethical approval. The protocol for this study was approved by the Siriraj Institutional Review Board (SIRB) of the Faculty of Medicine Siriraj Hospital, Mahidol University (certificate of approval no. Si394/2013, date of approval 8 July 2013). The protocol was also registered with ClinicalTrials.gov (reg. no. NCT02103764, date of approval 18 Feb 2014). The initial participant enrollment was on 11 March 2014. A written informed consent was obtained from all participants before enrollment. Received: 29 January 2021; Accepted: 10 January 2022

1. Munro, M. G., Critchley, H. O. D. & Fraser, I. S. The two FIGO systems for normal and abnormal uterine bleeding symptoms and classification of causes of abnormal uterine bleeding in the reproductive years: 2018 revisions. Int. J. Gynaecol. Obstet. 143, 393–408 (2018). 2. Fritz, M. A. Abnormal uterine bleeding. In Clinical Gynecologic Endocrinology and Infertility (eds Fritz, M. A. & Speroff, L.) 591–621 (Lippincott Williams & Wilkins, 2011). 3. Ferenczy, A. Pathophysiology of endometrial bleeding. Maturitas 45, 1–14 (2003). 4. Horbelt, D. V ., Roberts, D. K., Parmley, T. H. & Walker, N. J. Ultrastructure of the microvasculature in human endometrial hyper- plasia. Am. J. Obstet. Gynecol. 174, 174–183 (1996). 5. Abulafia, O. & Sherer, D. M. Angiogenesis of the endometrium. Obstet. Gynecol. 94, 148–153 (1999). 6. Hickey, M., Higham, J. M. & Fraser, I. Progestogens with or without oestrogen for irregular uterine bleeding associated with anovulation. Cochrane Database Syst. Rev. 9, CD001895 (2012). 7. Nilsson, L. & Rybo, G. Treatment of menorrhagia. Am. J. Obstet. Gynecol. 110, 713–720 (1971). 8. Sitruk-Ware, R. Progestins and cardiovascular risk markers. Steroids 65, 651–658 (2000). 9. Darney, P . D. The androgenicity of progestins. Am. J. Med. 98, 104S-110S (1995). 10. Stanczyk, F . Z. All progestins are not created equal. Steroids 68, 879–890 (2003). 11. Schindler, A. E. et al. Classification and pharmacology of progestins. Maturitas 46(Suppl 1), S7–S16 (2003). 12. Fraser, I. S. Treatment of ovulatory and anovulatory dysfunctional uterine bleeding with oral progestogens. Aust. N Z J. Obstet. Gynaecol. 30, 353–356 (1990). 13. Stone, S. C. Desogestrel. Clin. Obstet. Gynecol. 38, 821–828 (1995). 14. Kaplan, B. Desogestrel, norgestimate, and gestodene: The newer progestins. Ann. Pharmacother. 29, 736–742 (1995). 15. Trivedi, N., Chauhan, N. & Vaidya, V . Effectiveness and safety of dydrogesterone in regularization of menstrual cycle: A post- marketing study. Gynecol. Endocrinol. 32, 667–671 (2016). 16. Podzolkova, N., Tatarchuk, T., Doshchanova, A., Eshimbetova, G. & Pexman-Fieth, C. Dydrogesterone treatment for menstrual- cycle regularization in routine clinical practice: A multicenter observational study. Gynecol. Endocrinol. 32, 246–249 (2016). 17. Wang, L., Guan, H. Y ., Xia, H. X., Chen, X. Y . & Zhang, W . Dydrogesterone treatment for menstrual-cycle regularization in abnor- mal uterine bleeding - ovulation dysfunction patients. World J. Clin. Cases 8, 3259–3266 (2020). 18. Seif, M. W ., Diamond, K. & Nickkho-Amiry, M. Obesity and menstrual disorders. Best Pract. Res. Clin. Obstet. Gynaecol. 29, 516–527 (2015). 19. National Institute for Health Care and Excellence. Heavy menstrual bleeding, (London, 2007). 8 Vol:.(1234567890)Scientific Reports | (2022) 12:1662 | https://doi.org/10.1038/s41598-022-05578-0 www.nature.com/scientificreports/ 20. Committee on Practice, B.-G. Practice bulletin no. 128: Diagnosis of abnormal uterine bleeding in reproductive-aged women. Obstet. Gynecol. 120, 197–206 (2012). 21. World Health Organization. Regional Office for the Western Pacific. The Asia-Pacific Perspective: Redefining Obesity and Its Treat- ment, (Health Communications Australia, 2000). 22. Maxwell, G. L. et al. Progestin and estrogen potency of combination oral contraceptives and endometrial cancer risk. Gynecol. Oncol. 103, 535–540 (2006). 23. Beral, V ., Bull, D., Reeves, G. & Million Women Study, C. Endometrial cancer and hormone-replacement therapy in the Million Women Study. Lancet 365, 1543–1551 (2005). 24. Rattanachaiyanont, M. et al. Clinical and pathological responses of progestin therapy for non-atypical endometrial hyperplasia: A prospective study. J. Obstet. Gynaecol. Res. 31, 98–106 (2005). 25. Di Carlo, C. et al. Bleeding patterns during continuous estradiol with different sequential progestogens therapy. Menopause 12, 520–525 (2005). 26. Roland, M., Clyman, M. J., Decker, A. & Ober, W . B. Classification of endometrial response to synthetic progestogen-estrogen compounds. Fertil. Steril. 15, 143–163 (1964). 27. Jiang, Y . & Tian, W . The effects of progesterones on blood lipids in hormone replacement therapy. Lipids Health Dis. 16, 219 (2017). 28. Committee on Practice, B.-G. Practice bulletin no. 136: Management of abnormal uterine bleeding associated with ovulatory dysfunction. Obstet. Gynecol. 122, 176–185 (2013). 29. Saghaei, M. Random allocation software for parallel group randomized trials. BMC Med. Res. Methodol. 4, 26 (2004). 30. Fraser, I. S. et al. Improving the objective quality of large-scale clinical trials for women with heavy menstrual bleeding: Experience from 2 multi-center, randomized trials. Reprod. Sci. 20, 745–754 (2013).

We would like to thank Dr. Chulaluk Komoltri, DrPh, Office for Research and Development, Faculty of Medicine Siriraj Hospital, Mahidol University for comments on research methodology and statistical analysis, and Dr. Kevin P . Jones, Ph.D., Khon Kaen University for English language editing. We also thank Ms. Chongdee Dangrat, M.Sc. and all Gynecologic Endocrinology Unit staff members, Department of Obstetrics and Gynecology, for coordinating the project and taking care of the study participants, respectively. Author contributions N.S. and M.R. contributed to the research initiation and study design, analysis and interpretation of data, and writing the manuscript. M.W . contributed to histopathology examination and interpretation of data. T.W ., S.I., P .T., K.T. and S.A. contributed to data collection, interpretation of data, and drafting the manuscript. All authors have approved the submitted version of the manuscript. Funding The study was supported by the Research and Development Fund of the Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand. Competing interests The authors declare no competing interests. Additional information Correspondence and requests for materials should be addressed to M.R. Reprints and permissions information is available at www.nature.com/reprints. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. © The Author(s) 2022