Comparative analysis of low-dose letrozole versus GnRH agonist on implantation markers and IVF outcomes in symptomatic adenomyosis: a randomized trial

Baseline and IVF cycle characteristics were compared among control patients (Group 1, n  = 75) and those with adenomyosis undergoing either Letrozole treatment (Group 2, n  = 79) or GnRHa treatment (Group 3, n  = 77). Age and duration of infertility showed no significant differences among the groups ( p  = 0.86 and p  = 0.76, respectively). However, BMI and basal serum estradiol levels were significantly higher in the patients with adenomyosis ( p  = 0.02 and p  = 0.03 respectively), while AMH levels and endometrial thickness were similar across all groups ( p  = 0.50 and p  = 0.31) (Table  1 ). The distribution of embryo transfer on days 3 and 5 was not significantly different among the groups ( p  = 0.36 and p  = 0.64). Table 1 Baseline characteristics. Parameters Control ( n  = 75) (Group 1) Adenomyosis P value Letrozole treated ( n  = 79) (Group 2) GnRH-A treated ( n  = 77) (Group 3) Age (years) 33.36 ± 2.84 34.14 ± 3.43 33.42 ± 2.93 0.8681 Duration of infertility (years) 7.1 ± 3.47 8.39 ± 1.68 8.66 ± 1.64 0.7644 BMI (Kg/m 2 ) 24.77 ± 3.78 24.75 ± 3.56* 24.27 ± 3.55* 0.0212 Anti Mullerian Hormone (ng/ml) 2.23 ± 0.59 2.26 ± 0.59 2.17 ± 0.58 0.5080 Estradiol (pg/ml) 30.39 ± 3.86 32.87 ± 4.45* 32.74 ± 4.17* 0.0373 Dose of gonadotrophins (IU) 2087.17 ± 600.44 2303.59 ± 531.54 2416.91 ± 305.10 0.6034 Estradiol on day of HCG (pg/ml) 1865.51 ± 459.76 2025.28 ± 287.23 2011.32 ± 390.94 0.2515 Progesterone on day of HCG (ng/ml) 0.98 ± 0.57 1.07 ± 0.55 1.05 ± 0.61 0.7179 Number of MII oocytes retrieved 8.49 ± 1.72 6.80 ± 1.90 7.08 ± 1.46 0.6798 Endometrial thickness (mm) 11.96 ± 1.77 10.19 ± 1.94 10.72 ± 2.02 0.3152 Number of grade I embryos 4.1 ± 1.09 3.6 ± 1.52 3.9 ± 1.38 0.5329 Day 3 embryo transfer 45 (60%) 51 (64.56%) 51 (66.23%) 0.3615 Day 5 embryo transfer 30 (40%) 28 (35.44%) 26 (33.77%) 0.6427 * P value < 0.05 was considered significant. Baseline characteristics. * P value < 0.05 was considered significant. Pregnancy outcomes and obstetric complications were presented in Tables  2 and 3 respectively. The total pregnancy rate was not significantly different among the control (40%), Letrozole-treated (35.44%), and GnRHa-treated (36.36%) groups ( p  = 0.64). No significant differences were observed in biochemical pregnancy ( p  = 0.38) or clinical pregnancy rates ( p  = 0.25). The live birth rate was lower in both adenomyosis groups compared to controls (Letrozole 17.72%, GnRHa 19.48%, control 30.67%), though not statistically significant ( p  = 0.17). Miscarriage rates were comparable across groups ( p  = 0.77). Obstetric complications such as antepartum hemorrhage, intrauterine growth restriction, pre-eclampsia, and postpartum hemorrhage occurred at similar frequencies among all groups, indicating that adenomyosis and treatment choice did not significantly alter the risk of adverse maternal or fetal outcomes. Table 2 Pregnancy outcomes. Parameters Control ( n  = 75) Adenomyosis P value Letrozole treated ( n  = 79) GnRH-A treated ( n  = 77) Total pregnancy, n (%) 30/75 (40%) 28/79 (35.44%) 28/77(36.36%) 0.6465 Biochemical pregnancy (%) 2/75 (2.67%) 06/79 (7.59%) 05/77 (6.49%) 0.3824 Clinical pregnancy rate (%) 28/75 (37.33%) 22/79 (27.85%) 23/77 (29.87%) 0.2509 Live birth rate, n (%) 23/75 (30.67%) 14/79 (17.72%) 15/77 (19.48%) 0.1758 Miscarriage 4/28 (14.29%) 6/22 (22.73%) 5/23 (21.74%) 0.7789 Pregnancy outcomes. Table 3 Obstetric complications. Parameters Control ( n  = 75) Adenomyosis P value Letrozole treated ( n  = 79) GnRH-A treated ( n  = 77) Antepartum haemorrhage 5/75 (6.67%) 8/79 (10.13%) 7/77 (9.09%) 0.7372 Intrauterine growth restriction 4/75 (5.33%) 9/79 (11.39%) 8/77 (10.39%) 0.3782 Pre-eclampsia 4/75 (5.33%) 7/79 (8.86%) 6/77 (7.79%) 0.6928 Postpartum haemorrhage 7/75 (9.33%) 10/79 (12.66%) 8/77 (10.39%) 0.7933 Obstetric complications. To identify predictors of pregnancy outcomes, a logistic regression analysis was done (Table  4 ). None of the predictors were significantly associated with any pregnancy outcomes, as all confidence intervals crossed the OR threshold of 1. For total pregnancy, the ORs (Fig.  2 ) for age, BMI, DOI, estradiol, and AMH were 0.97, 0.97, 0.90, 0.98, and 0.82, respectively, with corresponding p -values indicating no meaningful impact. Similar non-significant results were observed for live birth, clinical pregnancy, and miscarriage. Additionally, clinical pregnancy and live birth rates were analyzed in patients with adenomyosis treated with Letrozole or GnRHa, stratified by lesion location. JZ (Group A) showed similar clinical pregnancy rates (10.12% and 10.38%) and live birth rates (5.06% and 3.89%) between the two treatment groups. The OM (Group B) had higher clinical pregnancy rates (16.45% for Letrozole and 19.48% for GnRHa) and relatively better live birth rates (12.65% and 14.28%) than JZ and advanced adenomyosis. Advanced adenomyosis (involvement of both JZ and OM; Group A + B) had lower pregnancy and live birth rates, indicating the negative impact of widespread disease. Logistic regression showed a significant association between lesion location and pregnancy outcomes, with a higher likelihood of pregnancy in patients with OM involvement compared to those with JZ lesions (Table  5 ). Fig. 2 Forest plot depicting the odds ratios (log scale) of key predictor variables—age, anti-Müllerian hormone (AMH), body mass index (BMI), duration of infertility (DOI), and estradiol—for four reproductive outcomes: total pregnancy, clinical pregnancy, live birth, and miscarriage. Forest plot depicting the odds ratios (log scale) of key predictor variables—age, anti-Müllerian hormone (AMH), body mass index (BMI), duration of infertility (DOI), and estradiol—for four reproductive outcomes: total pregnancy, clinical pregnancy, live birth, and miscarriage. Table 4 Logistic regression analysis showing the association between clinical predictors and four reproductive outcomes: total pregnancy, live birth, clinical pregnancy, and miscarriage. Odds ratio CI (95%) P value Lower Upper Total pregnancy (Intercept) 21.29 0.30 1628.95 0.16 AGE 0.97 0.89 1.06 0.51 BMI 0.97 0.90 1.05 0.43 DOI 0.90 0.77 1.06 0.22 ESTRADIOL 0.98 0.92 1.05 0.59 AMH 0.82 0.52 1.32 0.42 Live birth (Intercept) 1.73 0.01 233.24 0.83 AGE 0.98 0.89 1.09 0.74 BMI 0.97 0.88 1.05 0.43 DOI 0.87 0.72 1.05 0.15 ESTRADIOL 1.03 0.95 1.11 0.48 AMH 0.97 0.57 1.69 0.92 Clinical pregnancy (Intercept) 14.64 0.17 1294.01 0.24 AGE 0.98 0.90 1.08 0.70 BMI 0.97 0.90 1.05 0.45 DOI 0.91 0.77 1.08 0.29 ESTRADIOL 0.98 0.91 1.04 0.45 AMH 0.78 0.48 1.27 0.31 Miscarriage (Intercept) 16.74 0.00 79767.95 0.51 AGE 0.99 0.83 1.19 0.91 BMI 0.93 0.79 1.08 0.34 DOI 1.06 0.77 1.47 0.71 ESTRADIOL 0.91 0.79 1.03 0.13 AMH 0.69 0.29 1.69 0.40 BMI, body mass index; DOI, duration of infertility; AMH,  anti-müllerian hormone; CI (95%) , 95% confidence intervals. Logistic regression analysis showing the association between clinical predictors and four reproductive outcomes: total pregnancy, live birth, clinical pregnancy, and miscarriage. BMI, body mass index; DOI, duration of infertility; AMH,  anti-müllerian hormone; CI (95%) , 95% confidence intervals. Table 5 Logistic regression analysis of clinical pregnancy and live birth rates in Letrozole- and GnRH-treated adenomyosis patients, stratified by lesion location. Group Letrozole treated ( n  = 79) clinical pregnancy GnRH treated ( n  = 77) Clinical Pregnancy Letrozole treated ( n  = 79) Live Birth GnRH treated ( n  = 77) live birth Estimate Std. error z value p -value Odds ratio Junctional zone (Group A) 8/79 (10.12%) 8/77 (10.38%) 4/79 (5.06%) 3/77 (3.89%) – – – – – Outer myometrium (Group B) 13/79 (16.45%) 15/77 (19.48%) 10/79 (12.65%) 11/77 (14.28%) − 1.182 0.4581 2.58 0.00988 3.26 Advanced adenomyosis (Group A + B) 1/79 (1.26%) 0/77 (0%) 0/79 (0%) 1/77 (1.29%) – – – – – Intercept (A + A + B) – – – – − 2.8067 0.3432 -8.177 2.91E−16 - Logistic regression analysis of clinical pregnancy and live birth rates in Letrozole- and GnRH-treated adenomyosis patients, stratified by lesion location. ERα and ERβ were found to be considerably increased ( p  < 0.001) in both luminal and glandular epithelia of women with adenomyosis than in normal endometrium. Post letrozole or GnRHa therapy, documented a significant decrease in immunoreactivities in the stroma and glandular epithelium for both receptors. Lower expression of PR was observed in eutopic endometrium of adenomyosis than in controls. Post therapy, in both the treatment groups, intense immunoreactivity to PR was noted during the WOI, showing strong nuclear-cytoplasmic expression in the adenomyotic glandular epithelium (Fig.  3 ). Integrin immunolabelling was minimal before either treatment which was significantly ( p  < 0.001) increased post GnRHa therapy ( p  < 0.003) or letrozole ( p  < 0.04) and was found similar to control population. Both angiogenic factors (VEGF, eNOS) exhibited down-regulation ( p  < 0.001) after letrozole or GnRHa treatment documenting strong immunoreactivity in endometrial glands with a weaker expression in stromal cells. Specifically, immunoreactivities to eNOS were detected in the luminal and glandular epithelium of endometrium and the endothelium of vascular vessels. In contrast, VEGF expression was detected in glandular epithelium only (Fig.  3 ). Fig. 3 Assessment of endometrial protein expression in women with adenomyosis before and after treatment with letrozole or a GnRH agonist, compared to healthy controls (PL: pre-treatment with Letrozole; PG: pre-treatment with GnRH Agonist, LT: post treatment with Letrozole; GT: post treatment with GnRH Agonist; Ctrl: control). ( A – G ): Bar plots showing improvement in implantation markers, namely, ( A ) aromatase, ( B ) ERα, ( C ) ERβ, ( D ) PR, ( E ) integrin αVβ3, ( F ) VEGF, and ( G ) eNOS. (* p  < 0.001; # p  < 0.01. *Comparison between treatment/s; #comparison vs. control) ( A ) Aromatase; ( B ) ER-α, ( C ) ER-β, ( D ) PR, ( E ) eNOS, ( F ) αVβ3 integrin and ( G ) VEGF after treatment with either letrozole or GnRH agonist in women with adenomyosis in comparison with controls. Assessment of endometrial protein expression in women with adenomyosis before and after treatment with letrozole or a GnRH agonist, compared to healthy controls (PL: pre-treatment with Letrozole; PG: pre-treatment with GnRH Agonist, LT: post treatment with Letrozole; GT: post treatment with GnRH Agonist; Ctrl: control). ( A – G ): Bar plots showing improvement in implantation markers, namely, ( A ) aromatase, ( B ) ERα, ( C ) ERβ, ( D ) PR, ( E ) integrin αVβ3, ( F ) VEGF, and ( G ) eNOS. (* p  < 0.001; # p  < 0.01. *Comparison between treatment/s; #comparison vs. control) ( A ) Aromatase; ( B ) ER-α, ( C ) ER-β, ( D ) PR, ( E ) eNOS, ( F ) αVβ3 integrin and ( G ) VEGF after treatment with either letrozole or GnRH agonist in women with adenomyosis in comparison with controls. The changes in implantation markers, as assessed by iELISA, were consistent with the findings from IHC in the adenomyosis cohort. A significant decrease ( p  < 0.0002) in ERα, ERβ, eNOS, and VEGF, along with an increase in PR ( p  < 0.008) and αvβ3 integrin ( p  < 0.04) was observed after either regimen. However, the cohort treated with GnRHa showed greater differences ( p  < 0.003) in integrin levels. The decrease in ERα and ERβ after letrozole treatment leads to controlled endometrial proliferation, thus possibly arresting pain and bleeding. Unsurprisingly, an increase in PR and αvβ3 integrin after GnRHa suggests aid in blastocyst apposition and adhesion (Fig.  4 ). Fig. 4 ( A – G ): Semi-quantitative scoring of different implantation markers in endometrial tissue of women with adenomyosis before and after treatment with either Letrozole or GnRH agonist in comparison with controls. ( A ) aromatase, ( B ) ERα, ( C ) ERβ, ( D ) PR, ( E ) VEGF, ( F ) eNOS, and ( G ) integrin αVβ3. (* p  < 0.0001; @ p  < 0.01, # p  < 0.008; ## p  < 0.0001 #, # #comparison between treatment/s; *, @ comparison vs. control) ( A – G ): Semi-quantitative scoring of different implantation markers in endometrial tissue of women with adenomyosis before and after treatment with either Letrozole or GnRH agonist in comparison with controls. ( A ) aromatase, ( B ) ERα, ( C ) ERβ, ( D ) PR, ( E ) VEGF, ( F ) eNOS, and ( G ) integrin αVβ3. (* p  < 0.0001; @ p  < 0.01, # p  < 0.008; ## p  < 0.0001 #, # #comparison between treatment/s; *, @ comparison vs. control)

The randomized control trial continues our previous research 28 and aims to further compare the effect of low dose letrozole and GnRHa on implantation markers and reproductive outcomes. This study was conducted at the Institute of Reproductive Medicine, a tertiary-care hospital in Kolkata, India. The study was approved by the institute’s Internal Human Ethics Committee and conducted in accordance with relevant guidelines and regulations.It was registered with the Clinical Trials Registry-India (CTRI/2019/01/016919). Written informed consent was obtained from all participants. Couples seeking IVF treatment were screened for adenomyosis, and 156 symptomatic women with diffuse adenomyosis were enrolled based on strict inclusion and exclusion criteria. Diagnosis was confirmed if three or more following features were present: globular uterus, asymmetrical thickening of the myometrium, myometrial cysts, echogenic sub-endometrial lines and buds, sub-endometrial cysts, hyperechogenic islands, fan-shaped shadowing, trans-lesional vascularity on Color Doppler, adenomyoma, and irregular or interrupted junctional zone (JZ) 29 , 30 (supplementary file 1). Adenomyosis was confirmed using a 2D TVS (SonoAce R7, Samsung-Madison, using a 4–9 MHz vaginal probe), performed by the same operator during days 5–10 of the menstrual cycle within three months of initiating IVF treatment. Exclusion criteria included age > 40 years, women with fibroids, hydrosalpinx, large uterine size (> 12 weeks), congenital uterine abnormalities, and AMH levels below 1.1 ng/mL. Participants were randomly allocated into two treatment groups using a computer-generated randomization table. Pre-treatment endometrial sampling was conducted on day 20/21 of the cycle; tissues were preserved for subsequent immunohistochemistry (IHC) and iELISA at − 80 °C. (Fig. 1 ). All patients underwent a long protocol downregulation for IVF, followed by embryo cryopreservation (supplementary file2). Fig. 1 Immunohistochemical (IHC) analysis of endometrial expression of estrogen receptor alpha (ER-α), estrogen receptor beta (ER-β), vascular endothelial growth factor (VEGF), endothelial nitric oxide synthase (eNOS), progesterone receptor (PR), and αVβ3 integrin in women with adenomyosis before and after treatment with either letrozole or a GnRH agonist, compared to healthy controls. Immunohistochemical (IHC) analysis of endometrial expression of estrogen receptor alpha (ER-α), estrogen receptor beta (ER-β), vascular endothelial growth factor (VEGF), endothelial nitric oxide synthase (eNOS), progesterone receptor (PR), and αVβ3 integrin in women with adenomyosis before and after treatment with either letrozole or a GnRH agonist, compared to healthy controls. During the subsequent cycle, one group (group 1) received letrozole orally (Letroz, Sun Pharmaceutical, India) at a dose of 2.5 mg thrice weekly for three months. The other group (group 2) received the depot GnRHa (goserelin, Zoladex 3.6 mg, AstraZeneca, UK) administered subcutaneously at 3.6 mg monthly for three consecutive months. The endometrium was prepared using hormone replacement therapy (HRT) with estradiol valerate (EV), starting at 6 mg daily in divided doses, administered 2–3 days after menstruation and titrated up to 12 mg/day. Endometrial thickness was monitored on TVS until it exceeded 7 mm. Subsequently, 600 mg of micronized progesterone was administered intravaginally for 5 days. Endometrial tissues were obtained using an endometrial biopsy curette after 5 days of progesterone administration (Probet Gynetics, India). The tissue was immediately divided into two parts: one was embedded for histological analysis and IHC while the other (500 mg) was weighed and homogenized to extract total protein in 1X T-PER tissue extraction buffer (Thermo Fisher Scientific, USA). Bicinchoninic acid assay (Thermo Fisher Scientific, USA) was conducted to ascertain protein concentration of the homogenate according to Sherpa et al. The homogenate was subsequently kept at -80 °C till further analysis. The levels of aromatase, ERα and ERβ, progesterone receptors (PR), integrin αvβ3, VEGF, endothelial nitric oxide synthase (eNOS) were assayed by iELISA. Paired endometrial tissue was taken from the control group (group 1; n  = 75) for comparison purpose. For histological analysis, tissue was fixed with 4% ( w / v ) paraformaldehyde in PBS for 2 h at room temperature; embedded in paraffin and sectioned at a thickness of 5 μm. Slides were processed for routine staining with haematoxylin and eosin. The sections were examined and photographed under a light microscope (Carl Zeiss Inc., 442 Thornwood, New York). For IHC, formaldehyde fixed paraffin-embedded 3–5 μm tissue sections were deparaffinized and treated with 0.3% hydrogen peroxide for 15 min and incubated with RetrivergenA (pH 5.6) (BD Bioscience, Franlin Lakes, NJ). Subsequent blocking was done with 2.5% bovine serum albumin (BSA) for 90 min at 4 °C. Sections were incubated with primary antibodies anti-aromatase (Abclonal, Cummings Park, Woburn, MA, USA) anti- ERα; anti ERβ; anti-PR; anti-integrin αvβ3; anti-VEGF; and anti-eNOS (Santa Cruz Biotechnology, Dallas, TX, USA) at 1:50 dilution for 12 h at 4 °C. Subsequently, the sections were incubated using a pre-diluted biotinylated secondary antibody (Santa Cruz Biotechnology, Dallas, TX, USA) for 3 h. Labelled cells were observed utilizing diaminobenzidine (Mumbai, India) followed by counterstaining with Mayer’s hematoxylin and then mounted with n-propylgallate containing 50% glycerol. A bright-field microscope (Leica Microsystems, Wetzlar, Germany) was utilized to view the sections and images were subsequently captured. Semi-quantitative scoring of IHC slides was independently performed by two blinded observers to assess staining intensity. Immunostaining intensity was categorized as 1, 2, or 3, corresponding to weak, moderate, and strong staining, respectively. Additionally, scores were assigned based on the percentage of stained cells: 0 for 0%, 1 for < 10%, 2 for 11–50%, and 3 for 51–80%. The final score was calculated to comprehensively reflect both staining intensity and distribution. A final immunohistochemical score ranging from 0 to 12 was determined by multiplying the number of stained cells and intensity score. Plate (Maxisorp F96, Nunc, Denmark) wells were coated with 25 µg/mL respective protein/50 µL. After overnight incubation at 4 °C, plates were washed three times with 1X phosphate-buffered saline (PBS) (pH = 7.2–7.6) and blocked with 2% BSA for 1.5 h at 37 °C. An optimal blocking buffer was chosen amongst 1% BSA, 2% BSA, 5% non-fat milk, and 5% FBS based on OD 450. Post blocking, plates were washed thrice with PBST (PBS with 0.05% Tween 20). Target proteins were detected using antibodies against VEGF, eNOS, ERα, ERβ, PR, and Integrin αvβ3 (Santa Cruz Biotechnology, Inc, Dallas, TX, USA), and aromatase (Abclonal, Cummings Park, Woburn, MA, USA) diluted at 1:500 and added 100 µL/well, followed by 2 h of incubation at 37 °C. Plates were washed three times with PBST before the addition of HRP-conjugated secondary antibodies (Abclonal, Cummings Park, Woburn, MA, USA) at 1:5000 for 2 h at 37 °C. After three washes with PBST, the colorimetric reaction was developed using 3,3’,5,5’-Tetramethylbenzidine (TMB) substrate for 30 min at 37 °C in the dark. The reaction was stopped after 20 min by adding 1 M sulphuric acid as per Volk et al., 2021 (1) and measured at 450 nm using a microplate reader (Thermo Scientific, MULTISKAN Sky High). All patients from both the treatment arms underwent a long protocol for IVF, which included daily Buserelin (0.5 mg, subcutaneously) starting from day 20/21 of the cycle. Pituitary suppression was confirmed on day 2/3 of the cycle, and then recombinant FSH 150–300 IU. (Gonal F, Serono, Geneva, Switzerland) and/or HMG 150–300 IU (Menopure; Ferring; Germany) was given. Folliculometry was started on day 6, and gonadotropin doses were adjusted accordingly. An HCG trigger of 10,000 IU (Sifasi; Serum Institute of India; India) was administered when at least 2 of the follicles reached 17 mm or more in dimension. Oocyte retrieval was performed 34–36 h post-trigger. Embryos were cultured up to day 3 or blastocyst and subsequently frozen for transfer at a later date. FET was carried out following the HRT treatment as mentioned above in preparation of the WOI. Endometrial thickness was monitored on TVS until it exceeded 7 mm. Subsequently, 600 mg of micronized progesterone was administered intravaginally in divided doses over 5 days for blastocyst or for 3 days for day 3 embryos and grade I embryos (maximum of two), were transferred based on availability. Luteal support commenced on the day of embryo transfer with daily intravaginal progesterone gel (90 mg; Crinone gel; Merck Serono; Italy). Serum βHCG levels were measured 14 days after embryo transfer, and if positive, luteal support continued up to the 12th week of pregnancy. TVS was performed around the 6th to 7th week of gestation to determine fetal viability. The primary outcome was live birth. Secondary outcome measures were biochemical pregnancy, clinical pregnancy, and miscarriage per embryo transfer cycle. The pregnancy rate was defined by the number of positive hCG (> 20 IU/L) assays 12 days after blastocyst transfer and 14 days after day 3 embryo transfer. Pregnancy was defined by the visualization of a gestational sac after 6 weeks of gestation. The clinical pregnancy rate was defined as the presence of a gestational sac and a live fetus on TVS at the 7th week of gestation. The live birth rate was defined as the delivery of a live fetus beyond 26 weeks of gestation. The target sample size was estimated according to Sharma et al. 2023 with an estimated power of 80% and an acceptable alpha error of < 0.05 and a beta error of 0.2. The sample size was 79 for the letrozole group and 77 for the GnRH agonist group in the sample size calculator ( https://clincalc.com/Stats/SampleSize.aspx ). The dataset was divided into Control (Group 1), Letrozole-treated (Group 2), and GnRHa-treated (Group 3) groups, with independent variables including age, body mass index (BMI), duration of infertility (DOI), estradiol, Anti-Mullerian Hormone (AMH), gonadotropin, estradiol on day of hCG, endometrial thickness (ET), progesterone on day of hCG, number of grade I embryos, and number of MII oocytes treated as continuous predictors. Dependent variables, e.g. total pregnancy, clinical pregnancy, live birth, and miscarriage were analyzed as binary outcomes. Descriptive statistics (mean ± SD) were computed for each independent variable within groups. Chi-square and Fisher’s exact tests were used depending on expected frequencies, with odds ratios (ORs) and 95% confidence intervals (CIs) computed via the Wald method. Separate logistic regression models were fitted to evaluate predictors of clinical pregnancy and live birth, reporting ORs, CIs, and p-values. Adenomyosis location was categorized into JZ (Group A), OM (Group B), and Advanced Adenomyosis (Group A + B), with clinical pregnancy and live birth rates calculated per group. Further comparisons assessed if Group B had higher pregnancy rates than the others, applying Fisher’s exact test and logistic regression models for outcome estimation. All analyses were performed using R statistical software with packages readr (data import/export), dplyr (data manipulation), base stats (logistic regression), broom (model outputs), and epitools (ORs and confidence intervals).

Adenomyosis has become increasingly prevalent among infertile women, a trend attributed to advancements in high-resolution ultrasound imaging and the rising age of women seeking fertility treatment. Its presence is widely acknowledged as a contributor to adverse reproductive outcomes. Nevertheless, the absence of randomized controlled trials directly comparing treatment modalities remains a significant limitation in addressing adenomyosis-associated infertility. This randomized study evaluated the impact of different pre-treatment strategies on IVF outcomes in symptomatic women diagnosed with adenomyosis. Specifically, we examined the efficacy of low dose letrozole and GnRHa, comparing their outcomes with a control group of women without the condition. Our findings demonstrated that both treatment regimens yielded comparable clinical pregnancy rates, live birth rates, and miscarriage rates in IVF cycles undergoing FET (Table  4 ). However, adenomyotic lesions localized to the JZ were significantly associated with reduced rates of clinical pregnancy and live birth compared to lesions in the OM. This highlights the particularly detrimental impact of JZ involvement on IVF outcomes. To elucidate the molecular mechanisms underlying these clinical observations, we assessed implantation markers in the eutopic endometrium using IHC and iELISA. Our analysis revealed aberrant expression of key markers, including estrogen receptors, progesterone receptors, integrin αvβ3, VEGF, eNOS, and aromatase, in women with adenomyosis. Pre-treatment with either low dose letrozole or GnRHa may improve these alterations, potentially enhancing endometrial receptivity and implantation potential. The existing literature consistently reports diminished implantation rates and suboptimal clinical outcomes in women with adenomyosis undergoing IVF 19 . Numerous studies, including several meta-analyses, have highlighted its adverse effects on clinical pregnancy, live birth, and miscarriage rates 17 – 19 . Conversely, some studies have reported no significant impairment in IVF outcomes 20 – 22 . This heterogeneity may be attributed to inconsistencies in diagnostic criteria, variation in disease severity classification, and differences in adenomyosis subtypes. Our ongoing investigation into adenomyosis and its clinical parameters has yielded several noteworthy findings. Women with adenomyosis exhibited significantly higher BMI compared to those without the condition, in agreement with previous studies 31 , 32 . These women with high BMI also reported increased dysmenorrhea and were more likely to present with ovarian endometriomas, corroborating findings by Hashim et al. (2020) 28 , 33 . The observed association between adenomyosis and elevated BMI may be explained by increased peripheral estrogen production via enhanced aromatase activity in adipose tissue. While most published literature reports normal systemic estrogen levels with elevated local concentrations in adenomyosis, our study observed significantly higher baseline serum estradiol levels in women with adenomyosis compared to controls, which may have important implications for the disease pathophysiology. Despite this elevation in basal estradiol, no significant difference was found between groups in the number of oocytes retrieved. Although prior studies indicate that basal estradiol levels below 20 pg/mL or above 80 pg/mL may increase the risk of cycle cancellation 34 , if the retrieval cycles are successful, these levels do not appear to influence pregnancy or delivery outcomes. In addition, we observed that endometrial thickness in adenomyosis patients was significantly reduced compared to controls, a finding consistent with Li et al. (2021) 35 . However, this reduction did not adversely affect pregnancy rates, as the mean endometrial thickness remained above the clinically accepted threshold of 7 mm, considered optimal for implantation and pregnancy 36 . Multivariate logistic regression analysis revealed no statistically significant effect of elevated BMI, increased baseline estradiol levels, or reduced endometrial thickness on overall reproductive outcomes in our study cohort. The association between symptomatic adenomyosis and reduced clinical pregnancy and live birth rates in IVF cycles is well established. Clinical symptoms such as dysmenorrhoea and menorrhagia may serve as indicators of more severe disease, further compromising fertility outcomes 37 – 39 . Moreover, an increased number of disease features has been correlated with lower live birth rates, suggesting their potential utility as markers of disease severity 40 , 41 . In our cohort of patients, who had symptoms of adenomyosis and showed at least three key features of the condition, tended to have lower pregnancy outcomes as compared to controls. Adenomyosis frequently coexists with endometriosis, with an overlap reported in approximately 65 to 90% of cases 6 , 42 —a trend also reflected in our cohort. Previous studies indicate that the coexistence of adenomyosis and endometriosis adversely affects reproductive outcomes 43 . Interestingly, Alson et al. noted that the direct features of adenomyosis with concomitant endometriosis had higher live birth rates compared to women without endometriosis 40 . Our findings also demonstrates that adenomyotic lesions confined to the OM in such cases were associated with better clinical pregnancy and live birth rates than lesions involving the JZ. In our study, women with symptomatic adenomyosis exhibited reduced expression of progesterone receptors and integrin αvβ3, alongside elevated levels of estrogen receptors (ERα and ERβ), aromatase, VEGF, and eNOS. These molecular alterations correlated with symptom severity and impaired endometrial receptivity, contributing to suboptimal reproductive outcomes. Notably, we observed a marked overexpression of ERs in the eutopic endometrium relative to healthy controls, characterized by a predominance of ERβ and a decreased ERα/ERβ ratio—findings consistent with existing literature 44 – 46 . Overexpression of aromatase and ERβ has been implicated as a central mechanism in the pathophysiology of adenomyosis and its association with dysmenorrhoea and infertility 12 , 47 . In our study, pre-treatment with low dose letrozole or GnRHa improved expression patterns of implantation markers within the eutopic endometrium, leading to clinical pregnancy and live birth rates comparable to those in the control group (Fig. 1 ). Post-treatment analysis demonstrated a significant reduction in both ERα and ERβ expression in stromal and glandular compartments, potentially explaining the associated symptomatic relief 28 . Furthermore, the aberrant overexpression of aromatase—widely recognized as a risk factor in adenomyosis—was significantly attenuated following treatment ( p < 0.001). We also confirmed elevated VEGF expression in the adenomyotic endometrium, in line with previous findings 48 . Treatment with either GnRHa or letrozole significantly reduced both VEGF and eNOS expression, thereby enhancing implantation rates. Given the pivotal role of integrins in embryo adhesion, the observed dysregulation of integrin αvβ3 expression in adenomyosis likely contributes to poor pregnancy outcomes 49 . Our findings concur with those of Xiao et al. (2013), as we documented improved integrin expression following treatment with both agents 13 . Adenomyosis presents notable challenges in clinical management due to the absence of approved medical therapies and a paucity of high-quality evidence to inform treatment protocols. GnRHa remains a commonly employed intervention aimed at symptom relief and enhancement of IVF outcomes 23 . These agents exhibit antiproliferative, anti-angiogenic, and anti-inflammatory properties, while also promoting apoptosis within ectopic foci 50 . Although prolonged GnRHa downregulation prior to FET has been associated with improved clinical pregnancy rates 25 , other studies have found no significant benefit 19 . In our study, no significant differences were observed in clinical pregnancy, miscarriage, or live birth rates between patients pre-treated with letrozole and those pre-treated with GnRHa for three months prior to FET. Numerous studies advocate the use of a freeze-all strategy to optimize outcomes in adenomyosis patients, as it enhances embryo-endometrial synchrony and mitigates the detrimental effects of supraphysiological hormone levels during controlled ovarian stimulation 51 . In our cohort (Table 2 ), FET performed after pre-treatment revealed no significant differences in clinical pregnancy, miscarriage, or live birth rate between treated adenomyosis patients and healthy controls. AIs have emerged as effective therapeutic options in the management of refractory endometriosis 52 . AIs inhibit the increased aromatase activity found in both normal and ectopic endometrial tissues, thereby reducing local estrogen synthesis 12 . Initial report on the use of AIs in combination with GnRHa in a young patient demonstrated successful management of abnormal uterine bleeding and reduction in uterine volume 27 . In a comparative study, Badawy et al. (2012) showed that letrozole was as effective as GnRHa in reducing uterine volume and ameliorating symptoms over a three-month treatment period 26 . Letrozole, a selective AI, lowers systemic and local oestrogen levels, thereby relieving symptoms and improving endometrial receptivity in adenomyosis 26 , 28 . Aromatase inhibition with letrozole reduces local estradiol synthesis within adenomyotic lesions, modulating estrogen-dependent paracrine signalling that regulates angiogenic factors (VEGF, eNOS) and implantation molecules such as integrin αvβ3. This helps restore the molecular environment necessary for embryo adhesion and implantation. In contrast, GnRHa suppresses the hypothalamic–pituitary–ovarian axis, inducing a reversible hypoestrogenic state that alleviates adenomyosis symptoms by inhibiting lesion growth and inducing apoptosis of endometrial cells. The 2.5 mg thrice-weekly letrozole regimen was chosen based on its pharmacokinetic properties—particularly its half-life of approximately 48 h—allowing sustained local aromatase inhibition while reducing systemic hypoestrogenism. This approach is supported by our previous clinical evidence demonstrating symptom reduction with favorable safety profiles 53 . This intermittent dosing demonstrated comparable efficacy and optimized local estrogen suppression with minimal systemic effects, potentially explaining observed clinical benefits and tolerability. Clinical pregnancy, live birth, and miscarriage rates in the letrozole group were statistically similar to those in the GnRHa group. Previous work by Cozzolino et al. (2022) reported improved outcomes with a combination of letrozole and GnRHa in patients with severe adenomyosis as only GnRHa-induced pituitary suppression does not fully inhibit local estrogen production 19 . Our subgroup analysis based on adenomyotic lesion localization (Table 5 ) revealed significantly lower pregnancy rates in patients with JZ involvement compared to those with OM lesions. In the logistic regression model, the intercept was negative, whereas OM involvement showed a significant positive coefficient (Estimate = 1.182, p = 0.0099; OR = 3.26), indicating higher pregnancy and live birth rates in OM-predominant cases. JZ pathology, which directly disrupts the endometrial–myometrial interface and coordinated uterine peristalsis, is likely more detrimental to implantation. In contrast, OM-predominant disease exerts less direct impact on endometrial receptivity. This trend was consistent across both treatment arms, underscoring the crucial role of lesion location in influencing reproductive outcomes. The reduced implantation potential in JZ involvement likely results from impaired endometrial receptivity and deficient decidualization—both critical for successful embryo implantation. This finding aligns with current literature showing that adenomyotic lesions localized to the JZ are associated with worse reproductive outcomes 54 ., irrespective of whether it presented alone or in conjunction with endometriosis. This study’s strengths include its prospective design, random patient assignment at recruitment, and longitudinal follow-up till live birth. Additionally, our research correlated reproductive outcomes with the anatomical location of adenomyotic lesions. The modest sample size and lower-than-expected event rates limited our power to identify small live birth differences, highlighting the need for larger multicentric studies. This limitation may affect the generalizability of the findings to a wider population. Furthermore, inclusion of coexisting endometriosis reflects clinical reality but precluded formal multivariate analysis.

This randomized controlled trial demonstrates that low-dose letrozole is as effective as GnRHa depot in women with symptomatic diffuse adenomyosis, showing comparable pregnancy, miscarriage, and live birth rates. Importantly, the study observed significant improvements in altered implantation marker expression post-treatment in both groups, enhancing endometrial receptivity. Furthermore, the involvement of the JZ in adenomyosis, alone or combined with the OM, significantly impacts clinical pregnancy and live birth rates during IVF. These findings highlight the critical role of adenomyosis localization in reproductive outcomes and underscore the necessity for tailored treatment strategies based on lesion characteristics.

Adenomyosis is an estrogen-dependent common uterine disorder affecting reproductive-aged women 1 . It is characterized by abnormal presence of endometrial tissue within the myometrium, causing localized or diffuse thickening of the uterine wall, dysmenorrhea, menorrhagia, infertility, and miscarriages 2 , 3 . Diagnosis of adenomyosis has evolved from post-hysterectomy histological examination to advanced imaging techniques, with high-resolution transvaginal sonography (TVS) and magnetic resonance imaging (MRI) with comparable efficacy 4 . TVS has become the primary diagnostic modality owing to its widespread availability and cost-effectiveness in comparison to MRI 5 . Adenomyosis shares etiological factors and clinical features with endometriosis and often coexists in up to 89.4% of cases 6 . However, isolated adenomyosis without endometriosis is observed in approximately 10% of women experiencing subfertility 7 . Furthermore, adenomyotic features are commonly identified in women with recurrent miscarriages, repeated implantation failures post In-vitro fertilization (IVF), and those undergoing IVF at advanced maternal ages 8 . The failure of successful implantation in adenomyosis is attributed to several factors such as altered uterine peristaltic activity, reduced endometrial receptivity due to various functional and molecular abnormalities, disrupted interactions between the embryo and endometrium and impaired decidualization 9 , 10 . Local hyperestrogenism exacerbates the condition by increasing estradiol synthesis 11 , 12 and elevated estradiol receptors (ER) activity 13 ,which disturbs the endometrial-myometrial interface and facilitates invasion of endometrial tissue into the myometrium. The combination of estradiol dominance and progesterone resistance may alter endometrial receptivity. Additionally research demonstrate that women with adenomyosis exhibit reduced αvβ3 v 3 integrin, elevated endothelial nitric oxide synthase (eNOS), and increased levels of vascular endothelial growth factor (VEGF) compared to controls 13 . The management of infertility associated with adenomyosis presents significant challenges for clinicians 14 , 15 . Although IVF remains the preferred treatment for adenomyosis related infertility 16 , numerous studies have reported adverse reproductive outcomes including higher rates of implantation failure and miscarriage 17 – 19 . Conversely, some investigations have found comparable reproductive outcomes in women with and without adenomyosis 20 – 22 . Gonadotropin releasing hormone agonist (GnRHa) is frequently used in adenomyosis to alleviate dysmenorrhoea and menorrhagia and to improve reproductive outcomes in the IVF cycle 23 . Prolonged downregulation with GnRHa before frozen embryo transfer (FET) has been shown to increase clinical pregnancy 24 , 25 ; However, the effects of GnRHa on endometrial receptivity in women with adenomyosis require further elucidation. Increased aromatase activity 12 in adenomyotic tissue has prompted use of aromatase inhibitors (AIs) for symptomatic relief. While limited studies have evaluated the efficacy of AIs in women with adenomyosis associated infertility 26 , 27 , literature also suggests that the use of letrozole has been utilized for endometrial preparation to enhance endometrial receptivity and clinical pregnancy rates. This prospective randomized controlled trial aimed to compare the effects of low dose letrozole and GnRHa on implantation markers and reproductive outcomes in this patient population.

Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1 Supplementary Material 2 Supplementary Material 2