Efficacy of progestogen therapy for extragenital endometriosis: clinical outcomes and MRI-detected changes in lesion characteristics

Endometriosis is a chronic inflammatory disease that affects approximately 10–15% of women of reproductive age and manifests predominantly as pain, including dysmenorrhea, dyspareunia, and chronic pelvic pain [2]. Persistent pain is an important predictor of increased anxiety and maladaptive responses, which may subsequently lead to depression and impair social and sexual functioning in women [3].

Laparoscopy remains the “gold standard” for the diagnosis of endometriosis; however, current guidelines place greater emphasis on non-invasive methods, particularly ultrasonography (US) and magnetic resonance imaging (MRI), which have high sensitivity and specificity for extragenital endometriosis (EGE) [6–8]. The advantages of MRI extend beyond timely and accurate diagnosis to include the dynamic assessment of lesion volume during therapy [9].

Clinical guidelines recommend progestogens, including dienogest and dydrogesterone, as first-line therapy [6]. Studies have demonstrated the predominantly local effects of dienogest on endometriotic lesions, with minimal systemic activity, as well as a significant reduction in lesion volume, particularly that of ovarian endometriomas (OE) [10–12]. A 2024 systematic review by Wu et al., which included 256 patients with various forms of deep endometriosis (DE), demonstrated that dienogest therapy significantly reduced pain severity and resulted in a statistically significant reduction in the size of DE lesions (p<0.00001) [13]. Most studies suggest that dienogest therapy primarily stabilizes the size and activity of DE and prevents its progression [13, 14].

Another progestogen with comparable clinical effects is dydrogesterone, which is structurally similar to endogenous progesterone [15–17]. In 2023, management algorithms for patients with endometriosis were developed that clearly outlined dydrogesterone treatment regimens for various clinical situations and highlighted its advantages in patients with reproductive plans [16]. A 2025 study demonstrated that 10 mg/day dydrogesterone reduced the intensity of dysmenorrhea by 75% as early as the third treatment cycle. In addition, the study reported a reduction in OE volume in 85% of the patients [18].

Despite the clinically established efficacy of hormonal therapy, objective criteria for assessing treatment outcomes using MRI remain insufficiently studied. In addition to lesion size, increasing attention is being paid to advanced functional MRI characteristics, including diffusion and perfusion parameters and signal intensity (SI) on T2-weighted (T2-W) images, as additional markers of lesion activity. SI correlates with the histological structure of the pathological lesion, depending on the relative proportions of the glandular and stromal components, extent of fibrosis, and degree of perifocal inflammatory response [19]. Therapeutic effects on endometriotic lesions may result in the remodeling of various components of the ectopic endometrium: the stromal component may become edematous, undergo a decidual-like reaction, and develop additional fibrosis and sclerosis [20]. The epithelial component may undergo secretory-like changes and become atrophic without evidence of functional activity. Thus, the SI of endometriotic lesions depends on both the structure of the individual lesion (predominance of glandular or stromal components, proliferative activity, and the extent of the fibrotic component and productive inflammation) and the therapeutic effect on the lesion and treatment duration.

This study aimed to evaluate the efficacy of dienogest and dydrogesterone in patients with external genital endometriosis with respect to the clinical course, quality of life (QOL) measures, and changes in the size and MRI signal characteristics of endometriotic lesions.

Materials and methods

This was a prospective, open-label, parallel-group, single-center study. Blinding was not performed because of the differences in the dosage forms and dosing regimens of the study drugs. The inclusion criteria were as follows: a confirmed diagnosis of EGE, including OEs, RCLs, or both OE and RCL, established by MRI; no absolute indication for surgical intervention and planned use of assisted reproductive technology programs at study enrolment; no previous surgical treatment for endometriosis; no use of sex hormone medications within 6 months before study enrolment; no contraindications to sex hormone therapy; and provision of written informed consent. The non-inclusion and exclusion criteria were as follows, malignant disease, severe somatic to understand the questionnaire, and no history of psychiatric disorders.

The primary endpoint was the change in EGE lesion volume (OE and/or RCL) on MRI after six months of treatment relative to baseline. The secondary endpoints included changes in lesion volume (OE and RCL) and MRI parameters (signal intensity on T2-weighted images), changes in quality-of-life measures assessed using the SF-20 and psychoemotional status assessed using the Hospital Anxiety and Depression Scale (HADS), and longitudinal changes in the prevalence of clinical symptoms (dysmenorrhea, dyspareunia, abnormal uterine bleeding, and cyclic abdominal bloating) and pain intensity assessed using a visual analog scale (VAS). Qualitative visual assessment of MRI scans to evaluate individual lesion dynamics was performed by a radiologist, and the cases were classified as showing a clear reduction in lesion size/volume, no substantial change in size (stable lesions), or a visually detectable increase in lesion size and volume. Quantitative statistical analysis was performed in all cases using continuous measurement of lesion volume. This approach is consistent with MRI-based studies assessing the efficacy of progestogens for endometriotic lesions [12–14, 18, 19], in which changes were evaluated according to the mean or median change in lesion volume.

The study was reviewed and approved by the Research Ethics Committee of the V. I. Kulakov National Medical Research Centre for Obstetrics, Gynecology and Perinatology (protocol No. 10, October 20, 2022).

Pelvic MRI was performed between days 5 and 10 of the menstrual cycle using GE Signa Excite 1.5-T and GE Signa Architect 3.0-T scanners. Lesion volume and signal intensity were assessed using T1-weighted images with and without fat suppression (FS), T2-weighted images acquired in three planes (sagittal, axial, and coronal), and diffusion-weighted imaging (DWI; b value 800 s/mm²) for OEs. The total lesion volume was assessed in mm³ for OE and RCL lesions and calculated using the ellipsoid formula: π/6×a×b×c, where π/6≈0.5236, and a, b, and c are diameters in mm. Patients were eligible if one of the maximal dimensions of the RCL did not exceed 7 mm.

For quantitative MRI analysis, regions of interest (ROIs) were delineated within the lesion structure, and the following parameters were calculated: Max, Min, and signal intensity values, which were used for the quantitative assessment of lesion homogeneity and structure; mean signal intensity (SI); standard deviation (SD), which characterizes variability in signal values within the ROI and reflects lesion heterogeneity; and the ADC on parametric maps, which reflects the degree of restriction of water-molecule diffusion within tissue and indicates viscosity in OEs.

Patients completed several psychometric instruments: the VAS; the SF-20 quality-of-life questionnaire, which assesses health status across six domains—physical functioning, role functioning, social functioning, mental health, health perception (scored from 0%, representing the worst possible status, to 100%, representing the best status), and pain (scored from 0%, representing the best status, to 100%, representing the worst status); and the HADS questionnaire, which assesses anxiety (HADS-A) and depression (HADS-D). Scores of 0–7 indicate no signs of anxiety or depression, scores of 8–10 indicate subclinical symptoms, and scores of 11 or higher indicate clinically significant anxiety or depression.

The study included 89 patients with endometriosis (mean age, 30.7 (6.8) years) who were followed up at the Department of Gynecological Endocrinology of the V.I. Kulakov National Medical Research Centre for Obstetrics, Gynecology and Perinatology, Ministry of Health of the Russian Federation. Following diagnostic confirmation and provision of informed consent, patients were allocated to two groups using a random number generator in Microsoft Excel to minimize systematic bias. At enrolment, all women were comparable with respect to age, presenting complaints, extent of lesions on MRI, and psychometric scale scores. Group 1 (n=45) received dydrogesterone 20 mg/day, in accordance with the algorithm for the management of patients with endometriosis [16]. Group 2 (n=44) received continuous treatment with of 2 mg/day. The minimum follow-up duration was 3 months, and the maximum duration at the time of data analysis was 6 months.

Statistical analysis

Data were collected and organized using Microsoft Office Excel 2016 spreadsheets. Statistical analyses and data visualization were performed using Jamovi version 2.3.15.0 and GraphPad Prism version 9.4.1 software. The normality of the distribution of quantitative data was assessed using the Shapiro–Wilk test. Normally distributed continuous variables are presented as mean (M) and standard deviation (SD), whereas non-normally distributed variables are presented as median (Me) and interquartile range [Q1; Q3]. Categorical data are presented as absolute frequencies and percentages. Baseline comparability between groups was assessed using the Mann–Whitney U test for quantitative variables and the Fisher’s exact test for categorical variables. Within-group changes were assessed using paired comparisons between baseline and 6 months: the Wilcoxon signed-rank test was used for quantitative variables and McNemar’s test was used for binary variables. Data at 3 months are presented descriptively without separate statistical comparisons because this intermediate time point was not a study endpoint.

To compare drug efficacy between groups, the individual change (Δ) for each quantitative outcome was calculated as the difference between the 6-month and baseline values (Δ = 6 months − baseline). Between-group comparisons of Δ were performed using the Mann–Whitney U-test. The results are presented as the difference in median Δ values with 95% confidence intervals (CIs) estimated using the Hodges–Lehmann method. Differences were considered statistically significant at p<0.05.

Results

The groups were comparable with respect to key clinical, demographic, and MRI parameters at the time of enrolment (Table 1).

Data on the patients’ clinical manifestations before enrolment and during treatment are presented in Table 2. Dysmenorrhea was the most common complaint, reported by the majority of women (77/89): 37/45 (82.2%) in group 1 and 40/44 (90.9%) in group 2. The next most common complaint was abnormal uterine bleeding (AUB) (Table 2), which was reported by almost 65% of patients overall in both groups (group 1, n=23/45; group 2, n=29/44); approximately 75% of these cases involved heavy menstrual bleeding (HMB). Cyclical abdominal bloating was reported by 53.3% (24/45) of patients in group 1 and 45.5% (20/44) in group 2. Dyspareunia was reported by 51.1% (23/45) and 47.7% (21/44) of the patients.

Both groups showed favorable regression of the clinical manifestations (Table 2). After 3 months of therapy, the prevalence of dysmenorrhea decreased by almost half, occurring in 42.2% and 50.0% of patients, respectively. By month 6, its prevalence had decreased by more than five-fold in group 1 (n=6) and almost ten-fold in group 2 (n=4). The reduction in pain symptoms in group 2 was partly attributable to the development of amenorrhea in most patients by month 6 of treatment (21/40). In the dydrogesterone group, the mean VAS score decreased from 5.6 (2.3) to 2.2 (1.5) (mean reduction, 3.4 points; 95% CI, 2.7–4.1; p<0.001); in the dienogest group, it decreased from 6.1 (2.5) to 2.0 (1.8) (mean reduction, 4.1 points; 95% CI, 3.3–4.9; p<0.001).

After 6 months of treatment, the frequency of all reported complaints decreased significantly in both groups, except for intermenstrual bleeding (IMB). In group 1, this symptom persisted in 4/7 (57.1%) patients, resolved in 3/7 (42.9%), and developed de novo during therapy in five patients. In group 2, after 6 months of therapy, IMB persisted in only 1 of 7 patients who had reported this complaint before treatment.

Table 3 presents the changes in the quantitative measures of the applied scales during treatment. Both drugs were effective in reducing pain severity, as reflected by a statistically significant reduction in VAS scores (p<0.001 for both groups).

Assessment of changes in psychoemotional status revealed divergent patterns between the groups. During dydrogesterone therapy, patients demonstrated a significant decrease in Hospital Anxiety and Depression Scale–Anxiety subscale (HADS-A) scores, from 8.4 (3.5) to 6.2 (3.4) (mean reduction, 2.2 points; 95% CI, 1.2–3.2; p<0.001), whereas Hospital Anxiety and Depression Scale–Depression subscale (HADS-D) scores remained stable throughout the observation period (6.7 (3.6) to 6.7 (3.3); p=0.914). Among patients receiving dienogest, HADS-A scores did not change significantly (7.3 (3.9) to 6.5 (4.5); p=0.064); however, HADS-D scores increased from 5.9 (4.0) to 9.1 (3.8) (mean increase, 3.2 points; 95% CI, 2.0–4.4; p<0.001). These changes were indirectly reflected in the trajectory of the “mental health” domain of the SF-20 questionnaire (p=0.16).

Before treatment, among patients receiving dydrogesterone (group 1, n=45), subclinical HADS-D scores were observed in 13/45 (28.9%) women, whereas clinically significant HADS-D scores were observed in 7/45 (15.6%) women. In the dienogest group (group 2, n=44), subclinical scores were observed in 14/44 (31.8%) patients and clinically significant scores in 6/44 (13.6%) patients. No statistically significant differences were identified between the groups before therapy (p>0.05). At month 6 of treatment, the proportion of patients with subclinical HADS-D scores (8–10 points) increased to 18/45 (40.0%) in group 1 and 16/44 (36.4%) in group 2. However, the nature of these changes differed between groups: in group 1, patients entering the category of subclinical HADS-D scores (8–10 points) were predominantly those with initially high HADS-D scores (>10 points), whereas in group 2, they were patients with HADS-D scores <7 before treatment. After 6 months of therapy, marked differences in the score trajectories were observed. In group 1, high HADS-D scores (>10 points) persisted in only 3/45 (6.7%) patients. In group 2, high HADS-D scores (>10 points) were identified in 10/44 (22.7%) patients at month 6 of therapy; notably, four patients with subclinical scores transitioned to this category.

Changes in MRI characteristics and lesion volumes during hormonal treatment are shown in Table 4. Six months of dienogest therapy (group 2) resulted in a reduction in ovarian endometrioma (OE) volume from 2536 to 1566 mm³, with a mean decrease of 970 mm³ (95% CI, 650–1290; p<0.001), and RCL volume from 403 to 263 mm³, with a mean decrease of 140 mm³ (95% CI, 95–210; p<0.001). An example of this pattern is shown in Figures 1 and 2, respectively. Similarly, dydrogesterone therapy (group 1) was associated with a significant reduction in OE volume from 2409 to 1754 mm³, with a mean decrease of 655 mm³ (95% CI, 420–890; p=0.016), whereas the reduction in RCL volume, from 422 to 342 mm³, was not statistically significant (p=0.08). When changes from baseline (Δ) were compared between the groups, statistically significant differences were identified only for T2-weighted (T2-W) signal intensity (Table 5).

On an individual basis, dienogest therapy was more frequently associated with favorable changes (37 cases, 82%), whereas a stable RCL lesion size was observed in eight cases (18%). During dydrogesterone therapy (Figure 3), favorable changes were observed in 24/45 patients (53%), stable RCL lesion size in 14/45 (31%), and unfavorable changes in 7/45 (16%). No adverse changes were observed during dienogest therapy. Notably, no evidence of deep infiltration into the adjacent bowel was observed in patients with disease progression. Therefore, further surgical treatment was not required, and conservative treatment was changed after 6 months.

Progestogen therapy was associated with a statistically significant reduction in the mean T2-W signal intensity (mean: from 471 to 296 in the dydrogesterone group, p=0.002; and from 569 to 241 in the dienogest group, p=0.002) and improved lesion homogeneity (SD: from 131 to 89, p=0.03; and from 139 to 75, p=0.006, respectively). In addition, OEs with a volume greater than 1800 mm³ before treatment had lower baseline apparent diffusion coefficient (ADC) values than cysts with a volume below 1800 mm³ (0.75 [0.11] vs. 1.09 [0.18], respectively; p<0.001). OE and RCL lesions with baseline ADC values >0.97×10⁻³ mm²/s demonstrated greater volume reduction during progestogen therapy.

Discussion

Treatment with both dienogest and dydrogesterone resulted in a statistically significant reduction in pain intensity, as assessed using the VAS (p<0.001 in both groups). In the present study, abdominal bloating (“endo belly”) was identified as one of the clinical manifestations of endometriosis [21] and was reported by 44/89 (49.4%) patients in this study. During treatment, bloating resolved in 22/44 (50.0%) patients, which may indicate a beneficial effect of hormonal therapy not only on endometriotic lesions but also on associated functional disturbances; however, this observation requires confirmation in future studies.

The difference in the incidence of IMB during treatment was noteworthy. In the dienogest group, the incidence of IMB decreased from 15.9% (7/44) to 2.3% (1/44) by month 6, probably owing to the development of amenorrhea in the majority of patients (21/40, 52.5%) receiving continuous treatment. In contrast, IMB developed for the first time during treatment in five patients in the dydrogesterone group. The initial dydrogesterone dose was 20 mg/day; when IMB occurred and/or persisted at month 3 of treatment, the dose was increased to 30 mg/day according to the algorithm [16], resulting in resolution of IMB in 3/7 (42.9%) patients. These findings are consistent with those of the published literature [16] and indicate the need for an individualized approach to dosing regimen selection.

In the context of current approaches to the long-term management of patients with endometriosis, the effects of treatment on women’s psychological well-being, in addition to pain, are particularly important. Our findings underscore the need to assess quality-of-life measures, including psychoemotional status, before initiating treatment. All patients with high HADS scores were evaluated by a psychiatrist, and the verified conditions corresponded to moderate depression and did not require pharmacotherapy. The proportion of patients with subclinical depression according to the HADS-D increased in both groups, reaching 18 (40.0%) and 16 (36.4 %) patients in groups 1 and 2, respectively. However, the patterns of change differed between the groups. In group 1, the transition to the subclinical range predominantly occurred among patients with HADS-D scores >10 before treatment initiation (clinically significant levels), reflecting a favorable change. In group 2, the increase in the proportion of patients with subclinical HADS-D depression was attributable to new cases among patients whose previous HADS-D scores were less than 7. In group 1, HADS-D scores >10 persisted in only three patients at month 6 of treatment, whereas in the dienogest group, such scores persisted in 10/44 (22.7%) patients. Thus, treatment with dydrogesterone was associated with a significant reduction in HADS-A scores, whereas no significant change in anxiety levels according to the HADS-A was observed with dienogest treatment; however, an increase in depression scores according to the HADS-D was noted. Differences in the metabolic pathways and pharmacodynamics of these drugs may explain the variability in psychoemotional responses during treatment [22–24]. These findings are consistent with the published evidence supporting mental health screening before and during dienogest treatment [23].

Given the direct visualization afforded by magnetic resonance imaging (MRI), its use in endometriosis enables detailed assessment of lesion size and characteristics not only at initial diagnosis but also for effective treatment monitoring by evaluating the degree of lesion regression during therapy. In our study, a reduction in OE volume was demonstrated in both groups (p<0.001). The volume of RCL decreased by 34.7% in group 2 compared to 19% in group 1; however, the difference was not statistically significant (p=0.08).

One important criterion for treatment efficacy is the reduction in T2-weighted signal intensity, accompanied by increased lesion homogeneity. After 6 months of treatment with dydrogesterone and dienogest, a significant decrease in the T2-weighted signal intensity (mean) and SD was observed, reflecting increased lesion homogeneity. Between-group comparisons showed that the reduction in signal intensity was significantly greater in group 2 (p<0.001), which may indicate more rapid changes in the lesion structure during dienogest treatment. This reflects a reduction in inflammatory and/or hemorrhagic components. A persistent, long-standing inflammatory response leads to fibrosis, whereas fibrous and smooth muscle components may be less responsive to hormonal therapy. Progestogens are known to induce atrophy of glandular components, reduce their secretory activity, decrease inflammatory infiltration, and reduce fresh hemorrhagic changes within lesions [20]. In some patients, the reduction of edema and the inflammatory component may cause endometriotic lesions to become denser, as reflected by MRI findings and characterized by stabilization of lesion size. In addition, a decrease in ADC indirectly indicates the formation of a less active but denser and more homogeneous lesion with reduced fluid components and edema [9].

In our study, patients receiving dienogest demonstrated a significant reduction in the size of endometriotic lesions and objective improvement in pain-related outcomes. Importantly, despite the observed increase in HADS scores and changes in psychoemotional status in this group, treatment adherence remained high in both groups. One possible explanation for this high adherence is the recognition of the clinical and instrumental efficacy of treatment as an important therapeutic outcome. In particular, MRI visualization of endometriotic lesion regression at month 6 of treatment may have contributed to improved psychoemotional well-being by reducing the fear of disease progression and fostering a sense of disease control.

This study has several limitations.

Limitations. Blinding was not performed because of the differences in pharmaceutical formulations and dosing regimens of the drugs. A formal sample size calculation was not performed during the study planning stage; the sample size was determined by patient availability during the recruitment period and was comparable to that of other similar studies [13, 14, 18]. The study findings were interpreted as demonstrating the efficacy of each drug in the parallel groups. These results require confirmation in studies with a more rigorous design and prospectively calculated sample sizes.

Conclusion

The findings of this study indicate that both dienogest and dydrogesterone are effective in treating endometriosis based on clinical characteristics and MRI findings. Both drugs resulted in a statistically significant reduction in pain severity, including dysmenorrhea and dyspareunia; however, their divergent effects on psychoemotional status and quality of life should be considered when selecting a treatment. The efficacy of hormonal therapy for EGE can be assessed not only based on clinical findings but also by MRI, including the measurement of lesion volume, T2-weighted signal intensity, and ADC.

Psychometric assessment of patients’ status before treatment, for example, using the HADS and other instruments, may become one component of the evaluation, enabling more precise selection of hormonal therapy and, when necessary, the timely initiation of psychotherapeutic support. Improving the quality of life should be a key objective in the management strategy for patients with endometriosis to facilitate a personalized approach that considers individual characteristics and patient-reported symptoms.