Endometrial Changes Associated with Mifepristone: A Review

Introduction

Mifepristone, a progesterone receptor (PR) antagonist, is widely utilized in gynecological practice for medical termination of early pregnancy and management of conditions such as uterine fibroids and endometriosis. In addition, combined with methotrexate, it can improve the ectopic pregnancy outcome which is explored in depth.^1,2 In recent years, its therapeutic scope has extended to treat Cushing’s syndrome, meningioma, mental illness (eg post-traumatic stress disorder, bipolar disorder) and other diseases, often necessitating doses exceeding the conventional range.^3–5 While the drug is generally well-tolerated, its side effects listed in the drug’s instructions include mild nausea, vomiting, dizziness, abdominal pain, and uterine bleeding. Of particular clinical concern is the emerging evidence of endometrial thickening associated with mifepristone use. Although historically considered a rare complication, this phenomenon has been increasingly reported in the context of its expanded applications. The potential for endometrial thickening to mimic malignancy underscores the need for vigilant surveillance and accurate differential diagnosis to distinguish it from endometrial cancer. Despite growing clinical awareness, the underlying mechanisms and risk factors remain incomplete. This review aims to synthesize existing evidence, identify critical gaps in the current literature and propose directions for future research in the endometrial effects of mifepristone.

The Effect of Mifepristone on the Endometrium

Mifepristone, a steroid PR antagonist, demonstrates a fivefold higher binding affinity for PR than progesterone (P). Despite its high receptor affinity, mifepristone acts as a pure competitive antagonist, failing to initiate activation and thereby blocking progesterone-mediated signaling. This mechanism effectively inhibits endometrial transition from the proliferative to the secretory phase. Concurrently, mifepristone downregulates endometrial expression of both PR and estrogen receptor (ER), amplifying its anti-transformative effects through dual receptor modulation.⁶ Furthermore, mifepristone-induced glandular androgen receptor expression enhancement may play a role in the anti-proliferation effect, while the specific mechanism is unclear.^7,8

Dose-dependent cellular responses to mifepristone have been well-characterized in vitro and vivo. At the endometrial level, escalating doses promote apoptosis, as evidenced by caspase-3 activation.^8–10 Notably, this dose-dependent relationship extends to the hypothalamic-pituitary-ovarian axis (HPOA). Specifically, low-dose regimens primarily affect the pituitary gland, while high-dose administration exerts dual inhibition at both the hypothalamus and pituitary levels, resulting in ovarian suppression.

Current evidence seems to overwhelmingly support mifepristone’s dose-dependent endometrial suppression, underpinning its therapeutic utility in conditions like endometriosis and uterine leiomyoma.¹¹ However, paradoxical cases of endometrial thickening during prolonged use have been reported, occasionally mimicking malignancy on imaging and histopathology.

Complex Biological Effects of Mifepristone on the Endometrium

In 2008, the National Institutes of Health (NIH) introduced the term “progesterone receptor modulator-associated endometrial changes (PAECs)” to describe the heterogeneous endometrial responses observed with mifepristone use.¹² However, some scholars have characterized these changes as exhibiting “estrogen-like” effects for the proliferation effect.^6,7,13

Extensive research has been conducted to elucidate the correlation between PAECs and mifepristone dosage and duration.⁶ For instance, a case series report published in 2005 documented varying degrees of endometrial thickening in patients receiving mifepristone at doses up to 400 mg/d, down to 5 mg/d, over 6 months,¹⁴ in which a large number of gray endometrial carcinoma-like tissues were found in some patients after laparotomy, but the pathological result was simple endometrial hyperplasia. A prospective study, by Fleseriu M et al, involving 35 women with Cushing’s syndrome taking mifepristone (300–1200 mg/d) for 24 weeks reported endometrial thickening in 10 patients (28.6%), 3 of them underwent curettage for unresolved endometrial thickening.³ However, a meta-analysis from 2013, exploring effects of mifepristone on uterine leiomyoma in premenopausal women, indicated a tenuous association between mifepristone doses and endometrial atypical hyperplasia but noted a marked difference in endometrial thickness between mifepristone (2.5mg/d, 5mg/d,10mg/d) and placebo groups, emphasizing the need for rigorous endometrial monitoring during treatment.¹⁵ Carbonell JL et al’s clinical researches were included and exhibited good homogeneity.^16–18 The thickness of the endometrium returned to normal 12 months after stopping the mifepristone.¹⁷ Subsequent research, such as a 2016 double-blind controlled trial found no significant differences in the incidence of PAECs after treating endometriosis women with doses of 2.5, 5, or 10 mg/d for 6 months, but the 5 mg/d group exhibited the lowest incidence (9.1%), while the 2.5 mg/d group demonstrated the highest (16.7%).¹⁹ In the same year, a study 36 women with uterine fibroids treated with 50 mg/w of mifepristone for 6 months reported that only 15% maintained normal or atrophic endometrium, with the remainder showing varying degrees of hyperplasia.²⁰ Collectively, mifepristone doses ≤10 mg/d are generally safe, with 5 mg/d potentially offering the optimal balance between efficacy and safety. Notably, a 2022 clinical report proposed the use of ultra-low dose mifepristone may be considered for patients who do not respond to conventional doses (50 mg/d), particularly perimenopause women with uterine fibroids.²¹ High-dose mifepristone is rarely used in gynecological practice, with its primary applications limited to short-term management of conditions such as meningioma or Cushing’s disease. In such cases, intermittent short-course progestin treatment may prove advantageous particularly when the anti-PRs effects of mifepristone are not the primary therapeutic goal.²²

A significant shortcoming of the existing studies is the lack of comprehensive analysis of hormone levels before, during, and after mifepristone administration. While some studies stratified patients based on menstrual cycle timing, others provided insufficient hormonal data.^19,23 In clinical practice, mifepristone is typically initiated within 7 days of menstrual onset, but the situation that taking the drug at the onset of symptoms cannot be ruled out, complicating the assessment of whether baseline hormonal levels influence PAECs development. This gap underscores the need for systematic studies to clarify the role of hormone status in PAECs pathogenesis.

However, a 120-day double-blind randomized controlled trial by Brown A et al in 2002 revealed that healthy women in Edinburgh who initiated mifepristone during the middle of the menstrual cycle or luteal phase exhibited endometrial thickening and divergent E₂ trends, regardless of dose (5 mg/d or 2 mg/d). In contrast, healthy women in Shanghai who started treatment during the early follicular phase showed endometrial thinning and a decline in E₂ level, with a positive correlation.²⁴ These findings suggest that the pre-treatment E₂ level, alongside such as race, steroids, and diet, may influence PAECs development.

A 2018 clinical trial highlighted that PAECs are more prevalent in older women, potentially due to the decline in P production that occurs after menopause. The study also identified a strong correlation between Ki67 expression (a marker of cellular proliferation) and the severity of PAECs pathology.²⁵

While multiple factors contribute to PAECs development, hormonal influences appear to play a central role which future research should prioritize.

PAECs

It is used to describe specific histological alterations observed in the endometrium of patients undergoing long-term treatment with PR antagonists. These changes are characterized by benign, non-physiological, and non-proliferative features. Importantly, PAECs are reversible, with histological normalization typically occurring following discontinuation of PR antagonist therapy.^26,27 The specific histological manifestations of PAECs may vary depending on the pharmacological agent used, as well as factors such as dose, duration of treatment, and individual patient characteristics.

Changes in Eutopic Endometrium

Macroscopically, ultrasound examinations reveal increased endometrial thickness in patients receiving long-term mifepristone therapy. Histologically, endometrial biopsies demonstrate cystic glandular dilatation without atypical cells, initially suggestive of “simple hyperplasia”. However, further examination often reveals secretory changes and other features inconsistent with hyperplasia. Currently, PAECs are characterized by: cystic hyperplasia accompanied by secretory changes; non-synchronous alterations in interstitial and glandular tissue (ie increased mitosis, multiple fibrosis, and “chicken foot” vascular patterns without thrombosis in the interstitial; a combination of mitosis, altered secretion, apoptosis and inactivity in the gland).²⁷

These findings suggest that mifepristone disrupts the typical overlapping hormonal dynamics of E and P, potentially leading to simultaneous E and P stimulation within individual glans. However, the precise mechanism underlying these changes remains unclear.²⁷

Studies investigating the distribution of PR and estrogen receptor (ER) following mifepristone treatment have reported variable patterns depending on treatment periods or menstrual cycle phase. For example, a trial administering 50 mg/d of mifepristone during the early follicular phase for 6 months found significantly increased ER levels in the endometrial stroma, with no significant changes in the interstitial PR or glandular ER/PR levels between the groups.²⁸ When administered during the luteal phase, non-synchronous phenomena persisted, but findings on PR immunostaining were inconsistent. Some studies reported a decline in PR levels,²⁹ while others observed an increase.³⁰ These discrepancies may reflect differences in study design, including dose and timing of administration.

Notably, a 2002 study demonstrated that mifepristone reduces E₂ levels while increasing ER expression, which is highly sensitive to E2.¹³ In the early follicular phase, where E₂ predominates, reduced E₂ levels result in endometrial thinning. Conversely, in the luteal phase, where P dominates, decreased E₂ levels may impair the P effect which is based on E₂, leading to endometrial thickening. These findings suggest that initiating mifepristone during the early follicular phase, with careful consideration of the ovarian cycle phase and E₂ levels, may mitigate endometrial changes and enhance safety.

While hormonal influences may be central to PAECs pathogenesis, additional mechanisms may contribute to endometrial changes. For instance, cytokines may directly affect glandular mitosis, potentially reducing proliferative activity. Further research is needed to elucidate these pathways and their clinical implications.

Despite concerns regarding mifepristone’s endometrial effects, the literature indicates a low incidence of atypical hyperplasia or endometrial cancer.¹² In a study of 11 Chinese women with long-term mifepristone exposure published in 2022, only one endometrial cancer was reported, characterized by stromal infiltration but low Ki67 expression and absent mitosis.¹¹ In 2018, a study found that mifepristone downregulates HAND and upregulates fibroblast growth factor 18 (FGF18), which is associated with endometrial thickening and endometrial cancer. In contrast, ulipristal acetate (UPA), another PR antagonist, did not exhibit these effects.³¹ In the same year, the molecular expression of PAECs was subjected to intensive investigation, which suggested that PAECs upregulate three genes (THY1, ADAM12, TN-C) that do not participate in the endometrial cancer signaling pathway.³² Instead, these genes appear to alter the tissue structure, thereby providing further evidence that mifepristone is safe. It has been suggested that the placement of a levonorgestrel intrauterine system (LNG-IUS) after two months may address the safety concerns.³³ However, it should be noted that the indications for LNG-IUS do have some overlap with those for mifepristone.

Mifepristone is not an entirely safe option when the duration is long or the dosage is large. Therefore, it should be considered only when a superior agent is unavailable. Increased endometrial thickness monitoring is necessary if patients must take mifepristone for a long duration (≥3 months). Initial monitoring begins during the first 3 months,^34–36 and the following monitoring time may be suitably extended based on specific circumstances. The discontinuation of mifepristone can improve endometrial thickness.^27,37 Patients with endometrial thickness who experience symptoms like bleeding (thickness usually >8 mm, even over 20 mm) or perimenopause are advised to undergo curettage.^17,38,39 LNG-IUS can be employed as an emergency measure or succedaneum.³³

Changes in Ectopic Endometrium

Historically, research on mifepristone-induced endometrial changes has primarily focused on the eutopic endometrium, with limited attention given to the ectopic endometrium. However, a recent study demonstrated that the ectopic endometrium in ovarian lesions of the majority of individuals (8 out of 11) displayed histopathological changes similar to those observed in the eutopic endometrium, including inactive glands and mild secretory changes, but the ectopic was characterized by more pronounced glandular congestion and less discernible secretory alternations.¹¹ This discrepancy can be explained by the distinct biological behaviors exhibited by the two entities. For example, increased expression and activity of 3β-hydroxysteroid dehydrogenase (3β-HSD) in the ectopic enhance the conversion of cholesterol to androstenedione, a precursor for E₂ synthesis. Higher androstenedione levels lead to increased E₂ production. Additionally, ectopic tissue reduced the expression of CYP11A1, the enzyme responsible for synthesizing P, and exhibits a deficiency in PR-B and decreased levels of PR-A¹², further impairing progesterone signaling.

Conclusions

In summary, prolonged use of mifepristone is potentially associated with the development of PAECs. Although numerous reports have not documented pathological diagnoses of atypical hyperplasia or endometrial cancer, the molecular mechanisms underlying these endometrial changes remain incompletely understood. Clinically, the use of mifepristone in young women during the early follicular phase at doses of ≤10 mg/d appears to be safe, although monitoring E₂ levels is advisable. However, clinicians should exercise caution when prescribing mifepristone, particularly when more effective therapeutic alternatives are available. In cases where endometrial thickening is detected via ultrasound during or after mifepristone treatment, a thorough differential diagnosis —including endometrial biopsy or curettage —is essential to rule out atypical hyperplasia or malignancy, especially for patients with bleeding symptoms or the perimenopausal period. Appropriate follow-up treatment should be considered, such as LNG-IUS, rather than resorting to hysterectomy due to suspicion of malignancy.

Despite the development of numerous novel pharmaceutical agents with the potential to replace mifepristone, it remains a cornerstone in research as a foundational PR antagonist. Critical research gaps persist, including a deeper understanding of population-specific variations in PAECs and the development of precision administration strategies to minimize the risk of PAECs while maximizing therapeutic efficacy. Addressing these gaps will require rigorous randomized controlled trials (RCTs) with large, diverse cohorts and long-term follow-up.