Abstract
Objective
To compare the efficacy of the levonorgestrel‐releasing intrauterine system combined with metformin versus oral progestin combined with metformin in fertility‐preserving treatment for patients with atypical endometrial hyperplasia and early endometrial cancer, and to analyze related factors affecting efficacy.
Methods
From January 2015 to December 2023, clinical data of patients with atypical endometrial hyperplasia or early endometrial cancer were retrospectively analyzed. Patients were divided into a study group and a control group based on treatment method. Complete remission rate, remission time, recurrence rate, pregnancy outcomes, and adverse reactions were compared between groups. Multivariate logistic regression was used to identify factors associated with complete remission, recurrence, and pregnancy.
Results
Post‐treatment endometrial thickness decreased in both groups compared with pre‐treatment levels, with a greater reduction in the study group than the control group. Compared with the control group, the study group demonstrated a higher complete remission rate at 3, 6, 9, and 12 months, shorter time to onset and time to complete remission, lower recurrence rate, higher pregnancy and live birth rates, and lower incidence of adverse reactions. Multivariate logistic regression analysis identified age 35 years or older, body mass index (BMI) of 28 or greater, polycystic ovary syndrome, and pathologic type of endometrial cancer as risk factors for complete remission. BMI of 28 or greater and pathologic type of endometrial cancer were risk factors for recurrence. Age 35 years or older and BMI 28 or greater were risk factors for pregnancy. BMI stratification analysis demonstrated that patients with lower BMI achieved better treatment outcomes.
Conclusions
This retrospective study preliminarily suggests that the levonorgestrel‐releasing intrauterine system combined with metformin shows a trend toward better treatment response in fertility‐preserving treatment, manifested by higher complete remission rates, shorter time to remission, lower recurrence rates, and favorable pregnancy outcomes. In addition, obesity, age of 35 years or older, polycystic ovary syndrome, and pathologic type of endometrial cancer were identified as factors influencing treatment response.
Keywords: atypical endometrial hyperplasia, early endometrial cancer, fertility‐preserving treatment, levonorgestrel‐releasing intrauterine system, metformin
1. INTRODUCTION
Endometrial cancer (EC) is one of the most common gynecologic malignancies, predominantly affecting postmenopausal women. Studies indicate that by 2040, it is projected to become the third most common cancer and the fourth leading cause of cancer death among women. 1 Atypical endometrial hyperplasia (AEH) is a precancerous condition of EC. Approximately 30% of AEH patients will progress to EC, while 20%–50% of AEH patients concurrently have EC. 2 In recent years, the incidence of EC and AEH has been increasing dramatically, showing a global trend toward younger age groups. Among women aged 20–44 years, the incidence of early EC is approximately 7.1%. 3 Surgery remains the primary treatment for early EC and AEH. However, women who have not completed childbearing have strong fertility preservation needs. Consequently, fertility‐sparing treatments for young AEH and early EC patients are gaining increasing attention. Currently, progestogen is the preferred choice for fertility preservation therapy. However, high‐efficiency oral administration often leads to adverse reactions such as severe nausea and vomiting, weight gain, and liver function impairment, resulting in poor patient compliance. 4 The levonorgestrel‐releasing intrauterine system (LNG‐IUS), a sustained‐release system for local progestogen delivery, can reduce the adverse effects associated with oral progestogen. Metformin, an insulin sensitizer, has been shown in recent studies to reduce recurrence rates and enhance progestin treatment efficacy when used for fertility‐sparing therapy in patients with early EC and AEH. 5 The present study aims to analyze the efficacy of the LNG‐IUS combined with metformin versus high‐efficiency progestin combined with metformin in fertility‐sparing treatment for patients with early EC and AEH. It also analyzes relevant risk factors to provide safe, feasible treatment options for these patients.
2. MATERIALS AND METHODS
2.1. Data source
A total of 65 patients who underwent fertility‐preserving therapy at Shenyang Women and Infants Hospital between January 2015 and December 2023 were retrospectively enrolled. All patients had a pathologic diagnosis of AEH (51 women) or early EC (14 women). Based on the type of fertility‐preserving therapy received, patients were divided into two groups: the study group (n = 39) received the LNG‐IUS plus metformin, while the control group (n = 26) received megestrol acetate (MA) plus metformin.
Women were included if they met the following inclusion criteria. (1) Age was 45 years or less and there was a strong patient desire to preserve fertility. (2) Pathologic findings from the first hysteroscopic curettage (requiring confirmation by two pathologists for final diagnosis) showed endometrial atypical hyperplasia or endometrioid adenocarcinoma, Stage G1, with no cervical involvement, no myometrial invasion, no extrauterine metastases, and no lymphovascular invasion. (3) Pelvic magnetic resonance imaging (MRI) confirmed tumor size and extent of involvement, consistent with FIGO (the International Federation of Gynecology & Obstetrics) Stage IA. (4) Progesterone receptor expression was positive. (5) There were no contraindications to LNG‐IUS, high‐efficiency progestin, or metformin therapy. (6) No other factors were causing infertility. (7) There were no contraindications to pregnancy. (8) There were no other malignant tumors in the breast or female reproductive system. (9) Compliance and follow‐up conditions were good. The exclusion criteria were: (1) other histologic types of EC, (2) poor histologic differentiation, (3) presence of metastatic lesions, and (4) concurrent malignancies in other sites.
The initial ethics approval (No. 201935) was obtained in June 2020 from the Medical Ethics Committee of Shenyang Women and Infants Hospital, originally for a prospective study design. During the course of the study, to make the research more comprehensive and complete, we decided to add the collection and analysis of retrospective case data. As the content of the retrospective component was consistent with that of the prospective component, and in accordance with the opinion of the hospital ethics committee, retrospective data collection could be conducted under the same ethics approval framework. The 65 patients included in the present study did not overlap with the prospective trial protocol (ChiCTR2200055624). To avoid potential confusion during the submission process and to ensure transparency of the study design, we separately applied for a dedicated retrospective study ethics approval (No. PJ‐SC‐2026‐005) before submission, confirming that the design and implementation of this retrospective study compliesd with ethical standards. Informed consent was obtained for the use of all patients' clinical data. Written informed consent was obtained from the patients for the publication of any potentially identifiable images or data included in this article. Written informed consent was obtained from all participants.
Collected information from patients' initial visits included age, body mass index (BMI; calculated as weight in kilograms divided by the square of height in meters), pathologic type, presence of diabetes or insulin resistance, hypertension, polycystic ovary syndrome (PCOS), fasting blood glucose, and obstetric history. Informed consent was obtained for all clinical data usage.
2.2. Treatment protocol and efficacy assessment
All 65 patients underwent hysteroscopic curettage for endometrial pathology assessment. Following hysteroscopic pathologic confirmation, the study group received oral metformin (500 mg twice daily) combined with intrauterine placement of an LNG‐IUS (Mirena) 3–7 days after the initial hysteroscopic curettage. In the control group, patients received oral metformin (500 mg twice daily) combined with megestrol acetate (160 mg once daily). Efficacy was evaluated via hysteroscopic endometrial biopsy pathology every 3 months, documenting treatment‐period endometrial pathology results and transvaginal pelvic color Doppler ultrasound findings. Data collection was completed through review of outpatient and inpatient medical records, outpatient follow‐up visits, and telephone follow ups.
For patients in the study group (those with LNG‐IUS placement), hysteroscopic endometrial biopsy during each 3‐month follow‐up was performed according to the following procedures. Hysteroscopy was first performed to observe the position of the device and the morphology of the uterine cavity. The LNG‐IUS was removed using an IUD removal device and after removal of the device, hysteroscopic endometrial biopsy was performed, and endometrial tissue was curetted for pathologic examination. Following the biopsy, an intrauterine device was reinserted during the same procedure: it was recommended to place a new LNG‐IUS; if the patient strongly requested to retain the original device, the removed device was disinfected with iodophor and reinserted into the uterine cavity. All procedures were completed within the same operative session to ensure that treatment continuity was not significantly affected. For patients in the control group (those receiving oral high‐dose progestin), hysteroscopic endometrial biopsy was performed directly without device removal or reinsertion.
The efficacy of drug therapy was assessed through histopathologic examination of the endometrial tissue obtained via hysteroscopic curettage. Specific pathologic criteria for efficacy determination are as follows. 6 For complete remission: post‐treatment histopathology shows secretory phase endometrium, proliferative phase endometrium, or other types of endometrium without abnormal proliferative changes. Partial remission is indicated by reduction in endometrial lesion grade. For patients with AEH, post‐treatment pathology shows hyperplasia without atypia. For patients with early EC, post‐treatment pathology shows endometrial hyperplasia without atypia or atypical hyperplasia. In stable disease pathologic findings post‐treatment are identical to those pre‐treatment. Disease progression is indicated by pathologic findings post‐treatment indicating more severe disease than pre‐treatment. Examples include AEH progressing to EC, or well‐differentiated endometrial adenocarcinoma progressing to G2, G3, or other EC subtypes. Disease recurrence was defined as after achieving complete remission of endometrial lesions, subsequent follow‐up reveals recurrence of EC. The efficacy timepoints were defined as (1) time to response: the interval from treatment initiation to pathologic partial response, and (2) time to complete response: the interval from treatment initiation to pathologic confirmation of complete response.
2.3. Follow‐up during fertility‐sparing therapy
Patients underwent quarterly follow‐up evaluations covering clinical symptoms, endometrial pathology, transvaginal ultrasound, and, when necessary, tumor marker testing, liver/kidney function tests, and pelvic MRI to assess disease progression. Follow up concluded in June 2025.
2.4. Monitoring of adverse drug reactions
Long‐term oral administration of high‐efficiency progestogen may cause numerous systemic side effects, including weight gain, acne, nausea, headache, dizziness, vomiting, depression, and increased thrombotic risk. When necessary, dose reduction or treatment regimen modification could be considered. During LNG‐IUS use, intrauterine device‐related adverse effects may occur, including vaginal spotting, pelvic inflammatory disease, ectopic pregnancy, intrauterine device displacement or expulsion, and perforation. Patients undergoing fertility preservation should undergo regular follow‐up evaluations, including liver and kidney function tests, coagulation studies, complete blood counts, tumor marker monitoring, and breast ultrasound examination.
2.5. Post‐cure management
Patients undergoing fertility‐sparing treatment who achieve two consecutive pathologic complete remissions may be referred to reproductive medicine for assisted reproductive technology (ART) if they have an urgent desire for pregnancy. ART approaches should be tailored to individual circumstances. Alternatively, natural conception may be attempted for 3–6 months; if unsuccessful within this period, ART should be pursued. For unmarried patients or those temporarily deferring pregnancy, long‐term maintenance therapy may involve intrauterine placement of an LNG‐IUS or cyclic low‐dose progestin oral medication. Follow‐up visits should occur every 3–6 months, including a detailed menstrual history review, pelvic ultrasound assessment of endometrial status, and hysteroscopic endometrial biopsy when indicated. Concurrently, patients should be guided to maintain healthy lifestyle habits, such as balanced nutrition and appropriate exercise, and undergo regular gynecologic examinations to ensure long‐term health. Additionally, psychological support is crucial for alleviating anxiety, improving treatment adherence, and enhancing quality of life. Comprehensive management should focus not only on therapeutic efficacy but also on overall patient health, providing a foundation for individualized treatment planning.
2.6. Statistical analysis
Statistical analysis of relevant data was performed using SPSS 27.0 software. Normality of measurement data was assessed using the K‐S method. Normally distributed data were expressed as mean ± standard deviation, and comparisons between groups were performed using the independent samples t‐test. For skewed distributions, median and interquartile range (IQR) were used. Comparisons between groups were performed using the Mann–Whitney U test. Count data were expressed as frequency and rate (%). Comparisons between groups were performed using the χ 2 test or Fisher exact test.
Multivariate logistic regression analysis was conducted by following these steps. First, variables with statistically significant differences (P < 0.05) were selected through univariate analysis, and in combination with clinical reality, variables that might be associated with the research outcome were included. All the variables to be included were subjected to collinearity diagnosis, and the results showed that the variance inflation factor of each variable was less than 5, indicating the absence of multicollinearity. On this basis, all the variables were simultaneously included in the multivariate logistic regression model using the Enter method (forced entry method) for analysis. Values of P less than 0.05 were considered statistically significant. When comparing three independent samples, if the overall test indicated statistical significance, post‐hoc comparisons using Bonferroni correction or the Kruskal‐Wallis method were required. The corrected significance level for Bonferroni correction is α = 0.05/3 ≈ 0.0167.
3. RESULTS
3.1. Comparison of general data between the two groups
The study group comprised 39 patients, including seven with early EC and 32 with AEH. The control group included 26 patients, comprising seven with early EC and 19 with AEH. Comparison of general clinical data between the two groups showed no statistically significant differences (P > 0.05) (Table 1).
TABLE 1.
Analysis of general data between study and control groups. a
| Baseline data | Study group | Control group | t/Z/χ2 | P |
|---|---|---|---|---|
| Age, year | ||||
| <35 | 30 (76.9) | 16 (61.5) | 1.785 | 0.182 |
| ≥35 | 9 (23.1) | 10 (38.5) | ||
| BMI | ||||
| <28 | 30 (76.9) | 15 (57.7) | 2.708 | 0.100 |
| ≥28 | 9 (23.1) | 11 (42.3) | ||
| Pathologic type | ||||
| AEH | 32 (82.1) | 19 (73.1) | 0.743 | 0.389 |
| Early EC | 7 (17.9) | 7 (26.9) | ||
| Combined polycystic ovary syndrome | 7 (17.9) | 7 (26.9) | 0.743 | 0.389 |
| Diabetes or insulin resistance | 8 (20.5) | 10 (38.5) | 2.510 | 0.113 |
| Hypertension | 6 (15.4) | 5 (19.2) | 0.005 | 0.946 |
| Menstrual irregularities | 16 (41.0) | 13 (50.0) | 0.508 | 0.476 |
| Comorbid gynecologic diseases | 13 (33.3) | 7 (26.9) | 0.301 | 0.583 |
| Abnormal gynecologic MRI findings | 12 (30.8) | 7 (26.9) | 0.112 | 0.738 |
| Infertility | 23 (59.0) | 16 (61.5) | 0.043 | 0.836 |
| Endometrial thickness at diagnosis, cm | ||||
| < 1 | 23 (59.0) | 18 (69.2) | 0.705 | 0.401 |
| ≥ 1 | 16 (41.0) | 8 (30.8) | ||
| Previous pregnancy history | ||||
| Yes | 16 (41.0) | 11 (42.3) | 0.011 | 0.918 |
| No | 23 (59.0) | 15 (57.7) | ||
| Previous delivery history | ||||
| Yes | 5 (12.8) | 7 (26.9) | 1.231 | 0.267 |
| No | 34 (87.2) | 19 (73.1) | ||
| Age at menarche, year | 13.00 (12.00–14.00) | 14.00 (12.00–14.25) | −1.931 | 0.053 |
| Cholesterol, mmol/L | 4.92 (4.44–5.23) | 4.74 (4.50–4.97) | −1.292 | 0.196 |
| Triglycerides, mmol/L | 1.66 (0.91–2.30) | 1.63 (1.05–2.27) | −0.275 | 0.784 |
| Uterine volume at diagnosis, cm3 | 74.03 ± 22.77 | 77.33 ± 33.37 | −0.441 | 0.662 |
| HDL‐C, mmol/L | 1.22 (1.11–1.39) | 1.22 (1.09–1.34) | −0.040 | 0.968 |
| LDL‐C, mmol/L | 2.89 ± 0.75 | 2.71 ± 0.59 | 1.055 | 0.295 |
| Fasting blood glucose, mmol/L | 5.80 (4.90–6.50) | 5.70 (5.18–6.40) | −0.281 | 0.778 |
| CA125, U/mL | 15.60 (11.50–19.60) | 15.70 (12.18–19.68) | −0.261 | 0.794 |
| CA199, U/mL | 8.50 (5.10–19.90) | 7.45 (5.38–9.63) | −0.630 | 0.529 |
Abbreviations: AEH, atypical endometrial hyperplasia; BMI, body mass index (calculated as weight in kilograms divided by the square of height in meters); EC, endometrial cancer; HDL‐C, high‐density lipoprotein cholesterol; LDL‐C, low‐density lipoprotein cholesterol; MRI, magnetic resonance imaging.
Data are presented as median (interquartile range), mean ± standard deviation, or number (percentage).
3.2. Comparison of changes in endometrial thickness before and after treatment between the two groups
Intergroup comparison showed no statistically significant difference in endometrial thickness between the two groups before treatment (P > 0.05). However, upon achieving complete remission, both groups exhibited varying degrees of endometrial thinning compared with pre‐treatment levels. The study group had a statistically significant reduction in endometrial thickness compared with the control group (P < 0.05). Also, the difference in endometrial thickness between pre‐ and post‐treatment measurements was greater in the study group than in the control group, with this difference being statistically significant (P < 0.05) (Table 2).
TABLE 2.
Changes in endometrial thickness before and after treatment in the study and control groups. a
| Group | At diagnosis, cm | At complete remission, cm | Difference after treatment, cm |
|---|---|---|---|
| Study group | 0.90 (0.80–1.13) | 0.60 (0.50–0.70) | 0.30 (0.20–0.40) |
| Control group | 0.90 (0.70–1.00) | 0.70 (0.60–0.95) | 0.20 (0.10–0.20) |
| Z | −0.084 | −2.006 | −2.941 |
| P b | 0.933 | 0.045 | 0.003 |
Data are presented as median (interquartile range).
P <0.05 indicates statistically significant differences.
3.3. Comparison of treatment outcomes between the two groups
During the 12‐month maintenance therapy, the study group achieved complete remission in 32 cases, with 13 cases (33.3%) at 3 months, 23 cases (59.0%) at 6 months, 30 cases (76.9%) at 9 months, and 32 cases (82.1%) at 12 months. The control group achieved complete remission in 18 cases, with 1 case (3.8%), 5 cases (19.2%), 12 cases (46.2%), and 15 cases (57.7%) achieving complete remission at 3, 6, 9, and 12 months, respectively. The study group demonstrated significantly higher complete remission rates than the control group at 3, 6, 9, and 12 months post‐treatment (P < 0.05). Although the response rate (97.4% vs. 84.6%) and complete remission rate (82.1% vs. 69.2%) in the study group were higher than those in the control group, the differences were not statistically significant (P > 0.05). However, the time to onset of response and time to achieve complete remission were significantly shorter in the study group than in the control group (P < 0.05). Regarding long‐term outcomes, the recurrence rate in the study group was significantly lower than that in the control group (P = 0.002), but the pregnancy rate (P = 0.011) and live birth rate (P = 0.045) were significantly higher than those in the control group. There was no significant difference in miscarriage rates between the two groups (P > 0.05). Regarding safety, the incidence of adverse reactions was significantly lower in the study group than in the control group (P = 0.036). Subgroup analysis of pregnant patients showed that among patients who achieved pregnancy, there was no statistically significant difference in the distribution of pregnancy modes between the two groups (natural pregnancy: 33.3% versus 41.7%; ART pregnancy: 66.7% vs. 58.3%; P = 0.879). The time from complete remission to pregnancy was shorter in the study group than in the control group (10.5 ± 1.6 months vs. 13.1 ± 1.6 months), with a statistically significant difference (P < 0.001) (Table 3).
TABLE 3.
Analysis of treatment outcomes in the study group and control group. a
| Index | Study group | Control group | χ 2/Z/t | P |
|---|---|---|---|---|
| Complete remission | ||||
| 3 month | 13 (33.3) | 1 (3.8) | 8.026 | 0.005 b |
| 6 month | 23 (59.0) | 5 (19.2) | 10.049 | 0.002 b |
| 9 month | 30 (76.9) | 12 (46.2) | 6.460 | 0.011 b |
| 12 month | 32 (82.1) | 15 (57.7) | 4.623 | 0.032 b |
| Therapeutic effect | ||||
| Effective | 38 (97.4) | 22 (84.6) | 2.031 | 0.154 |
| Complete remission | 32 (82.1) | 18 (69.2) | 1.444 | 0.229 |
| Partial remission | 6 (15.4) | 4 (15.4) | ||
| Stable disease | 0 | 3 (11.5) | ||
| Disease progression | 1 (2.6) | 1 (3.8) | ||
| Time to onset of effect, month | 3.00 (3.00–6.75) | 6.00 (5.25–9.00) | −2.591 | 0.010 b |
| Time to complete remission, month | 6.00 (3.00–9.00) | 9.00 (6.00–12.00) | −3.403 | <0.001 b |
| Recurrence | 3 (7.7) | 10 (38.5) | 9.231 | 0.002 b |
| Adverse reactions | 5 (12.8) | 9 (34.6) | 4.385 | 0.036 b |
| Pregnancy | 30 (76.9) | 12 (46.2) | 6.460 | 0.011 b |
| Miscarriage | 15 (38.5) | 8 (30.8) | 0.404 | 0.525 |
| Live births | 15 (38.5) | 4 (15.4) | 4.016 | 0.045 b |
| Subgroup analysis in pregnant patients | (n = 30) | (n = 12) | ||
| Mode of pregnancy | ||||
| Natural pregnancy | 10 (33.3) | 5 (41.7) | 0.023 | 0.879 |
| ART pregnancy | 20 (66.7) | 7 (58.3) | ||
| Time from CR to pregnancy, month | 10.5 ± 1.6 | 13.1 ± 1.6 | −4.713 | <0.001 b |
Abbreviations: ART, assisted reproductive technology; CR, complete remission.
Data are presented as median (interquartile range), mean ± standard deviation, or number (percentage).
P <0.05 indicates statistically significant difference.
3.4. Analysis of factors affecting complete remission, recurrence, and pregnancy outcomes
Multivariate logistic regression analysis revealed that an age of 35 years or greater (odds ratio [OR] 24.26, P = 0.023), BMI 28 or greater (OR 12.06, P = 0.020), concomitant PCOS (OR 21.58, P = 0.021), and histologic type EC (OR 172.72, P = 0.029) were risk factors affecting complete remission (Table 4). BMI of 28 or greater (OR 14.73, P = 0.006) and histologic type EC (OR 24.04, P = 0.023) were risk factors affecting recurrence, as shown in Table 5. Being aged 35 years or older (OR 9.53, P = 0.019) and having a BMI of 28 or greater (OR 20.40, P < 0.001) were risk factors affecting pregnancy outcomes (Table 6).
TABLE 4.
Analysis of factors influencing complete remission.
| Variable | β | SE | Wald | OR | 95% CI | P a |
|---|---|---|---|---|---|---|
| Age ≥ 35 y | 3.189 | 1.405 | 5.153 | 24.26 | 1.55–380.76 | 0.023 |
| BMI ≥ 28 | 2.490 | 1.070 | 5.415 | 12.06 | 1.48–98.27 | 0.020 |
| Combined polycystic ovary syndrome | 3.072 | 1.336 | 5.287 | 21.58 | 1.57–295.83 | 0.021 |
| Diabetes or insulin resistance | −1.476 | 1.375 | 1.152 | 0.23 | 0.02–3.38 | 0.283 |
| Pathologic type: EC | 5.152 | 2.366 | 4.742 | 172.72 | 1.67–17822.44 | 0.029 |
| Abnormal gynecologic MRI findings | −3.252 | 2.147 | 2.295 | 0.04 | 0.001–2.60 | 0.130 |
| Endometrial thickness ≥ 1 cm at diagnosis | −3.190 | 1.771 | 3.245 | 0.04 | 0.001–1.33 | 0.072 |
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by the square of height in meters); CI, confidence interval; EC, endometrial cancer; MRI, magnetic resonance imaging; OR, odds ratio; SE, standard error.
P < 0.05 indicates statistically significant differences.
TABLE 5.
Analysis of factors associated with recurrence.
| Variable | β | SE | Wald | OR | 95% CI | P a |
|---|---|---|---|---|---|---|
| Age ≥ 35 year | 0.809 | 1.066 | 0.576 | 2.25 | 0.28–18.15 | 0.448 |
| BMI ≥28 | 2.690 | 0.979 | 7.546 | 14.73 | 2.16–100.35 | 0.006 |
| Combined polycystic ovary syndrome | 0.474 | 1.149 | 0.170 | 1.61 | 0.17–15.28 | 0.680 |
| Diabetes or insulin resistance | −1.683 | 1.073 | 2.461 | 0.19 | 0.02–1.52 | 0.117 |
| Pathologic type: EC | 3.180 | 1.397 | 5.180 | 24.04 | 1.56–371.68 | 0.023 |
| Abnormal gynecologic MRI findings | 0.417 | 1.035 | 0.162 | 1.52 | 0.20–11.54 | 0.687 |
| Endometrial thickness ≥1 cm at diagnosis | −0.991 | 0.965 | 1.055 | 0.37 | 0.06–2.46 | 0.304 |
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by the square of height in meters); CI, confidence interval; EC, endometrial cancer; MRI, magnetic resonance imaging; OR, odds ratio; SE, standard error.
P <0.05 indicates statistically significant differences.
TABLE 6.
Analysis of factors affecting pregnancy.
| Variable | β | SE | Wald | OR | 95% CI | P a |
|---|---|---|---|---|---|---|
| Age ≥ 35 year | 2.255 | 0.964 | 5.466 | 9.53 | 1.44–63.08 | 0.019 |
| BMI ≥ 28 | 3.016 | 0.884 | 11.645 | 20.40 | 3.61–115.30 | <0.001 |
| Combined polycystic ovary syndrome | −0.189 | 1.000 | 0.036 | 0.83 | 0.12–5.88 | 0.850 |
| Diabetes or insulin resistance | 0.641 | 1.071 | 0.359 | 1.90 | 0.23–15.48 | 0.549 |
| Pathologic type: EC | −0.270 | 1.325 | 0.041 | 0.76 | 0.06–10.26 | 0.839 |
| Abnormal gynecologic MRI findings | 0.422 | 1.013 | 0.173 | 1.53 | 0.21–11.11 | 0.677 |
| Endometrial thickness ≥1 cm at diagnosis | 1.156 | 0.992 | 1.356 | 3.18 | 0.45–22.22 | 0.244 |
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by the square of height in meters); CI, confidence interval; EC, endometrial cancer; MRI, magnetic resonance imaging; OR, odds ratio; SE, standard error.
P <0.05 indicates statistically significant differences.
Patients were grouped based on BMI: normal group (BMI < 24), overweight group (24 ≤ BMI < 28), and obese group (BMI ≥ 28). Treatment efficacy was compared across groups. Results showed that although the 3‐month complete remission rates among the three groups (34.3% versus 10.0% versus 5.0%) differed, these differences were not statistically significant. The normal BMI group exhibited significantly higher complete remission rates at 6 and 9 months post‐treatment compared with the obese group (P < 0.0167). The 12‐month complete remission rates in both the normal BMI and overweight groups were higher than those in the obese group, with statistically significant differences (P < 0.0167). Overall, the complete remission rate was higher in the normal BMI group than in the obese group, with a statistically significant difference (P < 0.0167). Regarding treatment duration, the time to response and time to complete remission were shorter in the normal BMI group than in the obese group, with statistically significant differences (P < 0.05). The relapse rate was highest in the obese group (40.0%), but the difference was not statistically significant. The pregnancy rate in the normal BMI group (94.3%) was significantly higher than in the other two groups (40.0%, 25.0%), with statistically significant differences between groups (P < 0.001). The live birth rate also showed a significant downward trend with increasing BMI. The live birth rates in the normal BMI and overweight groups were 45.7% and 30.0%, respectively; there were no live births in the obese group. The differences between groups were statistically significant (P = 0.002) (Table 7).
TABLE 7.
Analysis of treatment outcomes by body mass index group. a
| Index | BMI < 24 | ≤24 ≤ BMI < 28 | BMI ≥ 28 | χ 2/H | P b |
|---|---|---|---|---|---|
| Complete remission | |||||
| 3 month | 12 (34.3) | 1 (10.0) | 1 (5.0) | 0.024 c | |
| 6 month | 22 (62.9) | 3 (30.0) | 3 (15.0) | 12.712 | 0.002 |
| 9 month | 31 (88.6) | 6 (60.0) | 5 (25.0) | 22.606 | <0.001 |
| 12 month | 30 (85.7) | 9 (90.0) | 8 (40.0) | 15.130 | <0.001 |
| Complete remission | 32 (91.4) | 9 (90.0) | 9 (45.0) | <0.001 c | |
| Time to response, month | 3.00 (3.00–6.00) | 6.00 (3.00–9.00) | 9.00 (3.00–9.00) | 9.181 | 0.009 |
| Time to complete remission, month | 6.00 (3.00–9.00) | 9.00 (6.00–12.00) | 9.00 (6.00–12.00) | 9.119 | 0.041 |
| Recurrence | 3 (8.6) | 2 (20.0) | 8 (40.0) | 0.017 c | |
| Pregnancy | 33 (94.3) | 4 (40.0) | 5 (25.0) | 29.854 | <0.001 |
| Live births | 16 (45.7) | 3 (30.0) | 0 | 12.861 | 0.002 |
Abbreviation: BMI, body mass index (calculated as weight in kilograms divided by the square of height in meters).
Data are presented as median (interquartile range) or as number (percentage).
P < 0.0167 after Bonferroni correction indicates statistically significant differences.
Application of Fisher exact test.
4. DISCUSSION
The incidence of AEH and early EC is rapidly increasing among women of reproductive age. AEH and early EC primarily result from imbalances in endogenous and exogenous estrogen and progesterone levels, disrupting the endometrial proliferation and shedding cycle and leading to abnormal endometrial growth. Endogenous factors include anovulatory abnormal uterine bleeding, functional ovarian tumors such as granulosa cell tumors with estrogenic activity, and estrogen exposure without progesterone antagonism due to ovulatory dysfunction in PCOS. Exogenous factors primarily involve postmenopausal hormone replacement therapy and hormone‐related treatments for women with estrogen receptor‐positive breast cancer, such as tamoxifen, which can stimulate abnormal endometrial hyperplasia and subsequent malignant transformation. Clinically, EC patients often present with diabetes, obesity, and hypertension—collectively termed the “triad of endometrial cancer”—classifying it as a metabolic‐related tumor. 7 Patients often present without obvious clinical symptoms early. As the disease progresses, irregular vaginal bleeding and discharge, amenorrhea, and infertility may occur. The standard clinical treatment for AEH and early EC is hysterectomy with bilateral salpingo‐oophorectomy. However, this standard approach renders young women infertile, significantly impacting their quality of life. Preserving fertility is the preferred outcome for young patients with AEH or early EC who desire future children, making fertility‐sparing treatment strategies a critical research focus in gynecologic oncology.
Patients with AEH and early EC typically achieve favorable outcomes with standardized fertility preservation therapy. Common clinical approaches include oral progestogen (such as medroxyprogesterone acetate and megestrol acetate) and insertion of the LNG‐IUS (Mirena). Other hormone‐related therapies include gonadotropin‐releasing hormone‐α agonists, oral contraceptives, tamoxifen, and letrozole. In the present study, the intervention group received the LNG‐IUS combined with metformin, and the control group received megestrol acetate combined with metformin. Results showed that the study group achieved significantly higher complete remission rates at 3, 6, 9, and 12 months post‐treatment compared with the control group, with a markedly shorter time to complete remission (6.00 months versus 9.00 months). This therapeutic difference can be further explained from the perspectives of drug pharmacokinetics and local mechanisms of action. The LNG‐IUS resembles a “T” shape, made of polyethylene material with dimensions of 32 mm in both length and width. This device contains 52 mg of levonorgestrel, which is released into the uterine cavity at a rate of 20 μg/24 h. Under the influence of progesterone, the endometrium atrophies and thins, rendering it unsuitable for implantation. This mechanism provides contraceptive effects and can also be used to treat menorrhagia and protect the endometrium from continuous estrogen exposure. Although 10% of LNG is released into the bloodstream upon entry into the uterine cavity, a low concentration of LNG persists in the bloodstream. This affects the hypothalamic–pituitary‐ovarian axis, inhibiting ovulation and suppressing the synthesis of estrogen receptors in the endometrium. It reduces continued estrogen stimulation of the endometrium, promoting endometrial cell apoptosis and glandular atrophy, thereby inhibiting endometrial proliferation. Its effective duration is approximately 5 years. 8 Furthermore, the LNG‐IUS delivers exceptionally high local progestin concentrations to the uterine mucosa. Compared with oral progestins, it maintains lower systemic blood concentrations. Through localized release of a potent progestin, the LNG‐IUS minimizes hormone‐related adverse effects and optimizes tolerability in women, representing a significant advantage. Research by Goh et al. 9 also indicates that women receiving the LNG‐IUS achieve higher complete remission rates compared with those given oral progestin. In contrast, while oral megestrol acetate can achieve complete remission rates exceeding 80% in most patients, 10 , 11 it exhibits significant peak‐to‐trough fluctuations in blood concentrations. Additionally, as a result of the first‐pass effect in the liver, the actual drug concentration reaching the endometrium is limited. More critically, long‐term oral progestin use induces numerous systemic adverse effects, including weight gain, acne, nausea, headache, dizziness, vomiting, depression, breast tenderness, and impaired liver/kidney function. It also elevates women's risk of thromboembolic events. 12 These adverse effects compromise patient compliance, thereby limiting overall therapeutic efficacy. Data from the present study also indicate that the incidence of adverse reactions in the control group was significantly higher than in the study group (34.6% versus 12.8%). These adverse reactions may directly impact patients' willingness and ability to continue treatment.
Epidemiologic studies have demonstrated that metformin exerts positive effects in both the prevention and treatment of tumors, reducing the recurrence rate of EC and improving survival rates among patients with this disease. 13 Therefore, all patients in the present study received combined therapy with metformin. Metformin primarily inhibits the growth of EC cells through two pathways. First, the adenosine monophosphate‐activated protein kinase (AMPK)‐dependent antitumor effects (direct pathway), in which metformin activates the AMPK signaling pathway, inducing tumor cell cycle arrest and apoptosis to inhibit tumor growth. Multiple laboratory studies on endometrial carcinoma cell lines and primary endometrial carcinoma cell cultures have also demonstrated that metformin suppresses estrogen‐induced endometrial growth via the mechanistic target of rapamycin (mTOR)‐AMPK pathway. Second, insulin‐dependent antitumor effects (indirect pathway), through which metformin acts on the liver and muscle tissues by inhibiting hepatic glucose production and promoting muscle glucose uptake. This reduces circulating glucose and insulin concentrations while enhancing insulin sensitivity. Consequently, it suppresses glucose glycolysis, diminishing the supply of substances for cellular synthesis and thereby inhibiting tumor growth. Previous studies indicate that in patients with early EC and AEH undergoing fertility‐sparing treatment, metformin therapy reduces disease recurrence rates and enhances the efficacy of progesterone therapy. Moreover, for patients with a poor response to progesterone monotherapy, combined metformin and progesterone therapy proves effective. 5 This improvement is also significant in non‐obese, insulin‐sensitive patients without hypertension or diabetes. 14 As most retrospective data originate from EC patients with concomitant diabetes, further research is needed to confirm metformin's efficacy in patients without diabetes and determine its optimal clinical dosage for oncology settings. Practical clinical application remains a long‐term endeavor.
The present study found more pronounced improvements in endometrial thickness during treatment in the study group. Endometrial thickness serves as a critical imaging indicator for assessing lesion severity and treatment response, with excessive hyperplasia often associated with more severe cellular atypia and higher carcinogenic risk. The reduction in endometrial thickness was significantly greater in the study group than in the control group, indicating that the LNG‐IUS exerts a role in controlling endometrial hyperplasia. This mechanism involves decreasing estrogen and progesterone receptor expression while promoting Fas expression, thereby sustainably inducing epithelial atrophy and effectively antagonizing estrogen‐induced endometrial proliferation. 15 From a histologic perspective, this improvement in endometrial thickness may signify the normalization of endometrial glandular architecture and decidualization of stromal cells. These alterations not only facilitate lesion reversal but also create a more favorable uterine environment for subsequent embryo implantation.
The present study systematically compared the efficacy of LNG‐IUS combined with metformin versus oral high‐efficiency progestin combined with metformin in fertility‐sparing treatment for patients with AEH and early EC. This retrospective study suggests that the LNG‐IUS plus metformin group demonstrated significantly higher complete remission rates at 3, 6, 9, and12 months, with shorter time to onset and time to complete remission, lower recurrence rate (7.7% vs. 38.5%), with higher pregnancy and live birth rates (76.9% vs. 46.2%, 38.5% vs. 15.4%), and lower adverse reaction incidence (12.8% vs. 34.6%). Hence, LNG‐IUS plus metformin demonstrated superior efficacy compared with oral high‐efficiency progestin plus metformin.
The present study also examined the impact of age, BMI, coexisting PCOS, histologic subtype, coexisting diabetes or insulin resistance, abnormal gynecologic MRI findings, and endometrial thickness at diagnosis on factors associated with complete remission, recurrence after cure, and pregnancy outcomes in two patient groups. Results indicate that an age of 35 years or older, a BMI of 28 or greater, concomitant PCOS, and EC histology are risk factors for achieving complete remission. BMI of 28 or greater and EC pathology were risk factors for recurrence; having an age of 35 years or greater and BMI of 28 or greater were risk factors for pregnancy outcomes. Notably, when patients were grouped by BMI, results showed that those with lower BMI demonstrated better response to pregnancy preservation therapy, indicating obesity's critical impact on treatment efficacy. From a pathophysiologic perspective, adipose tissue functions not merely as a passive energy storage organ but as an active endocrine organ. It creates a microenvironment conducive to endometrial hyperplasia by producing excessive estrogen, inflammatory mediators (e.g., tumor necrosis factor‐α, interleukin‐6), and adipokines. Specifically, aromatase in adipose tissue converts androstenedione into estrogens, leading to elevated local and systemic estrogen levels. This sustained estrogenic stimulation serves as a major driver for the initiation and progression of endometrial pathology. 16 Furthermore, during progestin therapy, high estrogen levels antagonize progestin effects, compromising treatment outcomes. Studies indicate that weight loss reduces systemic hormonal and inflammatory stimuli promoting endometrial hyperplasia and may synergize with intrauterine progestogen to induce tumor cell death. 17 Further research shows that patients achieving 3% weight loss during treatment exhibit significantly reduced recurrence risk. 18 Regarding pregnancy outcomes, obesity may impair fertility through multiple pathways: at the ovarian level, obesity‐induced insulin resistance and chronic inflammation may affect follicular development and oocyte quality; at the endometrial level, obesity‐related endocrine disorders may compromise endometrial receptivity; and at the systemic level, obesity may disrupt hypothalamic–pituitary‐ovarian axis function, impairing normal ovulation. Therefore, during fertility treatment, patients are advised to manage weight through dietary control, exercise, and metabolic management, with medication used when necessary to achieve weight loss goals. PCOS, another significant factor affecting treatment outcomes, shares overlapping pathophysiologic mechanisms with obesity while possessing unique characteristics. The hallmark hormonal features of PCOS—hyperandrogenism and elevated luteinizing hormone/follicle‐stimulating hormone ratios—can form a vicious cycle with insulin resistance, jointly promoting abnormal endometrial hyperplasia. Histologically, EC manifests not only as more pronounced cellular atypia but also as significant glandular disorganization and stromal infiltration. At the molecular level, EC typically accumulates more gene mutations and epigenetic alterations, which may affect the expression and function of drug targets. For example, mismatch repair‐deficient tumors harbor tumor suppressor gene mutations in the PIK3C/AKT/mTOR pathway. These mutations may promote progesterone resistance by altering progesterone receptor expression and activity, thereby impacting the efficacy of pregnancy‐sustaining therapy. 19 Furthermore, an age of 35 years or older serves as a risk factor for complete remission and pregnancy outcomes. Its mechanism likely correlates with suboptimal endocrine‐metabolic status, diminished immune function, reduced ovarian reserve, decreased endometrial receptivity, and increased rates of chromosomal abnormalities in oocytes among older patients. With advancing age, significant changes occur in the female endocrine system, such as fluctuations in estrogen and progesterone levels. 20 Estrogen promotes proliferation and differentiation of endometrial cells, while progesterone inhibits excessive hyperplasia. Hormonal imbalance in older patients can reduce endometrial responsiveness to progesterone, impairing treatment efficacy and increasing recurrence risk. Immunologic function typically declines with age, diminishing surveillance and clearance of cancer cells. 21 Therefore, older patients should exercise greater caution when selecting fertility‐sparing treatments. A comprehensive fertility assessment is recommended before treatment, considering factors such as patient age, ovarian reserve, tubal function, and male fertility, to predict pregnancy likelihood following fertility‐sparing therapy. The appropriate decision regarding fertility preservation not only provides predictive value for post‐treatment pregnancy outcomes but also facilitates early planning for conception and life after treatment.
The present study demonstrated significantly higher pregnancy rates (76.9% vs 46.2%) and live birth rates (38.5% vs 15.4%) in the study group compared with the control group. Among patients who achieved pregnancy, there was no statistically significant difference in ART utilization between the two groups (66.7% vs 58.3%, P = 0.879), suggesting that the higher pregnancy rate in the study group was not solely attributable to a higher rate of ART use. Further analysis showed that the time from complete remission to pregnancy tended to be shorter in the study group than in the control group (10.5 ± 1.6 months vs 13.1 ± 1.6 months), with a statistically significant difference (P < 0.001). This may be attributed to two key factors: First, the study group exhibited faster treatment response and higher complete remission rates, thereby securing more valuable time for fertility preservation—a particularly critical factor for older patients. Second, the LNG‐IUS group may have experienced earlier restoration of the endometrial environment, including endometrial thickness, blood perfusion, and expression of receptivity‐related factors. Collectively, these elements created more favorable conditions for embryo implantation. In addition, the incidence of adverse reactions in the study group was lower, and the patients' treatment compliance was higher. This reduced treatment interruptions caused by adverse drug effects, thereby ensuring the continuity and timeliness of the treatment. In clinical practice, it is generally believed that achieving complete remission earlier after palliative treatment is beneficial for securing a valuable time window for subsequent pregnancy, and starting pregnancy as early as possible can reduce the potential impact of recurrence risk on the reproductive outcome. During fertility‐sparing treatment for AEH and early EC, repeated endometrial curettage procedures can damage the basal layer of the endometrium, leading to intrauterine adhesions and/or thin endometrium, which reduce pregnancy success rates. 22 Studies confirm that intrauterine adhesions are an independent risk factor affecting pregnancy success rates after treatment for AEH and early EC, with an incidence rate of approximately 15.2%. 23 Therefore, fertility‐sparing treatment should avoid excessive diagnostic curettage and prioritize endometrial protection. Concurrently, efforts should be made to explore methods that reduce the frequency of curettage and minimize endometrial trauma. The intrauterine placement of an LNG‐IUS can increase uterine cavity volume, thereby helping prevent intrauterine adhesions. For patients with existing adhesions, hysteroscopic adhesiolysis combined with hormonal therapy may be employed to improve the endometrial environment. However, it should be noted that the efficacy of intrauterine adhesion treatment varies among individuals, with a relatively high postoperative recurrence rate. Therefore, close monitoring is essential during treatment. For patients with fertility needs, it is recommended to attempt pregnancy as soon as possible after achieving complete remission to minimize the potential impact of intrauterine adhesions on pregnancy. Following successful preservation therapy, prompt fertility assistance is generally recommended, making ART the most common approach post‐treatment. Fan Y et al. 22 found that the use of ART, reversal of pre‐pregnancy BMI to normal levels, and shorter complete remission duration were positively correlated with successful pregnancy. Based on the findings of the present study and previous literature, after achieving complete remission in fertility‐sparing treatment for AEH and early EC, it is recommended that ART be implemented as early as possible to achieve favorable pregnancy outcomes.
Hence, for young patients with AEH or early EC, appropriate fertility‐sparing treatment options may be selected after thorough discussion of the condition, explanation of benefits and risks, and informed consent. However, regular histopathologic evaluation and follow up are essential. Fertility‐sparing treatment is safe, effective, and feasible for young patients with well‐differentiated AEH or early EC. Throughout fertility‐sparing treatment, emphasis should be placed on regular assessment of endometrial histopathologic changes. Patients with suboptimal treatment outcomes or recurrence should be promptly transitioned to staging surgery to prevent disease progression.
The present study possesses several notable strengths. First, it precisely focuses on comparing the efficacy and safety of two mainstream combined regimens for fertility preservation: a local progestin‐releasing system versus systemic oral progestin. The use of metformin as a common background in both groups effectively controlled for the confounding influence of this synergistic agent. Second, moving beyond traditional pathologic response rates, the study established a relatively comprehensive evaluation framework. This framework encompassed short‐term efficacy indicators (complete response rate, time to response), long‐term oncologic outcomes (recurrence rate), reproductive outcomes (pregnancy rate, live birth rate), and safety profiles (adverse event incidence), thereby providing robust data for a holistic assessment of the overall benefit of the treatment strategies. Furthermore, through multifactorial logistic regression analysis, the present study systematically investigated the independent impact of various clinical factors, including age, BMI, and pathologic type, on complete response, recurrence, and pregnancy outcomes. This aids in identifying high‐risk populations and provides a preliminary basis for implementing individualized treatment and intensified management in clinical practice. Finally, the BMI stratification analysis clearly demonstrated the negative impact of obesity on treatment response, reinforcing the evidence‐based rationale for integrating weight management concurrently with fertility‐preserving therapy.
However, the present study also has several limitations. The primary limitation lies in its retrospective nature, where the completeness and accuracy of the data are inherently constrained by reliance on pre‐existing medical records. Retrospective research depends on historical clinical documentation, which limits the ability to adequately adjust for important confounding factors and may introduce information bias. Second, the unavoidable issue of selection bias must be acknowledged. Treatment group allocation was influenced by factors such as patient preference and cost, introducing inherent selection and confounding biases. Although baseline characteristics showed no statistically significant differences between the groups, unmeasured or unknown confounders (e.g., socioeconomic status, intensity of treatment motivation) could have systematically influenced the results. Third, constraints related to sample size and statistical power are evident. The total sample size (n = 65) was relatively small, which reduces statistical power. This may have caused some clinically important differences to fail to reach statistical significance. It also limits the stability of the multifactorial regression models; with multiple variables included, there is a risk of overfitting, leading to imprecise estimates with abnormally wide confidence intervals for the odds ratios of some factors. Fourth, there is ambiguity in attributing the study findings. As both groups received combination therapy with metformin, the study design cannot isolate or quantify the independent contribution of metformin to the observed outcomes. The independent contribution of metformin and its specific mechanism of action require further investigation. Fifth, the present study only recorded ART utilization data for patients who achieved pregnancy; it did not collect ART utilization information for patients who did not achieve pregnancy. Therefore, ART use could not be incorporated as a variable in the multivariate regression analysis. Future prospective studies are needed to further clarify the independent role of ART in pregnancy outcomes following fertility‐preserving treatment. Sixth, a significant depth limitation is its “blind spot” in the molecular era: the absence of tumor molecular subtyping data. This confines the conclusions to the histologic level and precludes answering the core precision medicine question of which molecular subtype of patients would benefit most from which treatment regimen, thereby limiting the contemporary relevance of its clinical guidance. Seventh, the study scope was insufficient in breadth. For a population whose goals are preserving fertility and quality of life, the impact of treatment on quality of life, menstrual patterns, and psychological well‐being is crucial. The lack of such data represents a significant gap in the assessment framework. Finally the follow‐up duration presents limitations. The median follow‐up time may be insufficient for robustly assessing the risk of long‐term recurrence (particularly several years after treatment cessation) and for evaluating long‐term reproductive outcomes (such as the sustained impact on ovarian reserve). The long‐term safety of fertility‐preserving treatments requires data from more extended follow up.
In conclusion, the present study systematically compared the efficacy of LNG‐IUS combined with metformin versus oral high‐efficiency progestin combined with metformin in fertility‐sparing treatment for patients with AEH and early EC. The results suggest that during the fertility‐sparing treatment process, the LNG‐IUS combined with metformin group demonstrated a better therapeutic response trend compared with the oral high‐dose progestin combined with metformin group, including higher complete remission rates, shorter remission times, lower recurrence rates, and better pregnancy outcomes and safety. Furthermore, multivariate analysis preliminarily identified age 35 years or older, BMI 28 or greater, concomitant PCOS, and EC as key factors influencing treatment outcomes, providing a reference basis for identifying high‐risk populations and developing personalized treatment strategies. The present study provides valuable evidence‐based medical insights for optimizing fertility preservation protocols in endometrial lesions. However, limitations including insufficient sample size, lack of molecular subtyping data, and inadequate follow‐up duration necessitate cautious interpretation and limit the generalizability of the conclusions. Future research should strive to conduct prospective, multicenter, large‐sample randomized controlled trials. An ideal study design would need to integrate tumor molecular subtyping, implement objective compliance monitoring, and systematically collect comprehensive data related to quality of life. The aim is to achieve individualized, precise, and optimized fertility‐preserving treatment for EC, thereby effectively controlling the disease while maximizing the protection of patients' fertility potential and long‐term quality of life.
AUTHOR CONTRIBUTIONS
PG and XZ were involved in the planning and design of the study. PG, JG, and YL were involved in literature search and data collection. PG, MH, XW, and XZ contributed to the writing of the article. All authors reviewed the results, contributed to the article and approved the submitted version.
FUNDING INFORMATION
This research is supported by the Shenyang Young and Middle‐aged Scientific and Technological Innovation Talents Support Program, Number RC210200.
CONFLICT OF INTEREST STATEMENT
The authors have no conflicts of interest.
ACKNOWLEDGMENTS
The authors wish to thank the women who participated in this trial and the staff who worked on the study.
Gu P, Gao J, Li Y, Huang M, Wang X, Zhao X. Efficacy and related factors of levonorgestrel‐releasing intrauterine system combined with metformin in fertility‐preserving treatment for atypical endometrial hyperplasia and early endometrial cancer: A retrospective cohort study. Int J Gynecol Obstet. 2026;175:396‐407. doi: 10.1002/ijgo.71037
[Correction added on 21 May 2026, after first online publication: The affiliations for Pengpeng Gu, Jialu Gao, and Yuxin Li have been updated in this version.]
DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/supplementary materials, and further inquiries can be directed to the corresponding author. All methods were performed in accordance with relevant guidelines and regulations.
REFERENCES
- 1. Eskander RN, Sill MW, Beffa L, et al. Pembrolizumab plus chemotherapy in advanced endometrial cancer. New England Journal of Medicine. 2023;388(23):2159‐2170. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Doherty MT, Sanni OB, Coleman HG, et al. Concurrent and future risk of endometrial cancer in women with endometrial hyperplasia: a systematic review and meta‐analysis. PloSone. 2020;15(4):e0232231. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Corzo C, Santillan NB, Westin SN, Barrientos Santillan N, Ramirez PT. Updates on conservative management of endometrial cancer. Journal of Minimally Invasive Gynecology. 2018;25(2):308‐313. [DOI] [PubMed] [Google Scholar]
- 4. Guan J, Chen XJ. The present status of metformin in fertility‐preserving treatment for atypical endometrial hyperplasia and endometrioid endometrial cancer. Frontiers in Endocrinology. 2022;13:1041535. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Gonthier C, Trefoux‐Bourdet A, Luton D, Koskas M. Fertility‐sparing management of endometrial cancer and atypical hyperplasia. Gynecologie, Obstetrique, Fertilite & Senologie. 2017;45(2):112‐118. [DOI] [PubMed] [Google Scholar]
- 6. Chen XJ, Yang JX, Wang HY. Recommendations for fertility‐sparing treatment and evaluation of atypical endometrial hyperplasia and early endometrioid carcinoma. Zhonghua Fu Chan Ke Za Zhi. 2019;54(2):80. doi: 10.3760/cma.j.issn.0529-567x.2019.02.002 [DOI] [Google Scholar]
- 7. Antonio R, Ursula C, Antonio T, et al. Mismatch repair deficiency specifically predicts recurrence of atypical endometrial hyperplasia and early endometrial carcinoma after conservative treatment: a multicenter study. Gynecologic Oncology. 2021;161(3):795‐801. [DOI] [PubMed] [Google Scholar]
- 8. Lang JH, Leng JH, Deng S, et al. Chinese expert consensus on the clinical application of the Levonorgestrel intrauterine release system. Zhonghua Fu Chan Ke Za Zhi. 2019;54(12):815. doi: 10.3760/cma.j.issn.0529-567x.2019.12.005 [DOI] [PubMed] [Google Scholar]
- 9. Goh CSY, Loh MJM, Lim WW, et al. A multi‐centre randomized controlled trial comparing megestrol acetate to levonorgestrel‐intrauterine system in fertility sparing treatment of atypical endometrial hyperplasia. Journal of Assisted Reproduction and Genetics. 2024;41:1‐10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Ouasti S, Bucau M, Larouzee E, Clement De Givry S, Chabbert‐Buffet N, Koskas M. Prospective study of fertility‐sparing treatment with chlormadinone acetate for endometrial carcinoma and atypical hyperplasia in young women. International Journal of Gynecology & Obstetrics. 2022;157(2):452‐457. [DOI] [PubMed] [Google Scholar]
- 11. Zhao XL, Du ZQ, Zhang X, Yao Z, Liang YQ, Zhao SF. Fertility‐preserving treatment in patients with early‐stage endometrial cancer. Medicine. 2021;100(48):e27961. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Koskas M, Azria E, Walker F, Luton D, Madelenat P, Yazbeck C. Progestin treatment of atypical hyperplasia and well‐differentiated adenocarcinoma of the endometrium to preserve fertility. Anticancer Research. 2012;32(3):1037‐1043. [PubMed] [Google Scholar]
- 13. Vicky M, MacKay H, Ray‐Coquard I, et al. Endometrial cancer (primer). Nature Reviews. Disease Primers. 2021;7(1):88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Yang BY, Gulinazi Y, Du Y, et al. Metformin plus megestrol acetate compared with megestrol acetate alone as fertility‐sparing treatment in patients with atypical endometrial hyperplasia and well‐differentiated endometrial cancer: a randomized controlled trial. BJOG: An International Journal of Obstetrics & Gynaecology. 2020;127(7):848‐857. [DOI] [PubMed] [Google Scholar]
- 15. Gallos ID, Krishan P, Shehmar M, Ganesan R, Gupta JK. LNG‐IUS versus oral progestogen treatment for endometrial hyperplasia: a long‐term comparative cohort study. Human Reproduction. 2013;28(11):2966‐2971. [DOI] [PubMed] [Google Scholar]
- 16. Onstad MA, Schmandt RE, Lu KH. Addressing the role of obesity in endometrial cancer risk prevention and treatment. Journal of Clinical Oncology. 2016;34(35):4225‐4230. doi: 10.1200/JCO.2016.69.4638 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. MacKintosh ML, Crosbie EJ. Obesity‐driven endometrial cancer: is weight loss the answer? BJOG. 2013;120:791‐794. [DOI] [PubMed] [Google Scholar]
- 18. Chen JY, Cao DY, Zhou HM, et al. GnRH‐a combined therapy for patients with atypical endometrial hyperplasia and endometrial cancer who failed oral progestin treatment. Chinese Journal of Obstetrics and Gynecology. 2021;56(8):561‐568. doi: 10.3760/cma.j.cn112141-20210603-00298 [DOI] [PubMed] [Google Scholar]
- 19. Tahiri A, Tekpli X, Satheesh SV, et al. Loss of progesterone receptor is associated with distinct tyrosine kinase profiles in breast cancer. Breast Cancer Research and Treatment. 2020;183(3):585‐598. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Li X, Fan Y, Wang J, et al. Insulin resistance and metabolic syndrome increase the risk of relapse for fertility preserving treatment in atypical endometrial hyperplasia and early endometrial cancer patients. Frontiers in Oncology. 2021;11:744689. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Gnainsky Y, Granot I, Aldo PB, et al. Local injury of the endometrium induces an inflammatory response that promotes successful implantation. Fertility and Sterility. 2010;94(6):2030‐2036. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Fan Y, Li X, Wang J, Wang Y, Tian L. Analysis of pregnancy‐associated factors after fertility‐sparing therapy in young women with early‐stage endometrial cancer or atypical endometrial hyperplasia. Reproductive Biology and Endocrinology. 2021;19(1):118. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Xiao ZR, Lu Q, Zhou R, et al. Analysis of pregnancy outcomes after fertility‐sparing treatment in patients with atypical endometrial hyperplasia and endometrial cancer. Chinese Journal of Obstetrics and Gynecology. 2020;55(12):857‐864. doi: 10.3760/cma.j.cn112141-20200613-00501 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The original contributions presented in the study are included in the article/supplementary materials, and further inquiries can be directed to the corresponding author. All methods were performed in accordance with relevant guidelines and regulations.