Randomised, triple-blind, placebo-controlled trial of atosiban in women with previous embryo implantation failure and abnormal uterine contractions undergoing single blastocyst-stage embryo transfer: a study protocol

Abstract

Introduction

Atosiban may confer therapeutic benefits to specific subpopulations in assisted reproductive technology. The Phase I Atosiban study indicated potential improvements in live birth rates among women with previous implantation failure undergoing frozen-thawed blastocyst transfer who exhibited abnormal uterine contractions, although these findings did not reach statistical significance. Therefore, further investigations are warranted to thoroughly elucidate the efficacy of atosiban and to evaluate whether uterine contractions can serve as a reliable biomarker for its targeted application.

Methods and analysis

This is a single-centre, randomised, triple-blind, placebo-controlled trial aiming to enrol 792 infertile women aged 20–40 years with a history of at least one previous embryo implantation failure and abnormal uterine contractions prior to single blastocyst-stage embryo transfer. Eligible participants will be randomly assigned in a 1:1 ratio to receive either intravenous atosiban or a placebo before embryo transfer. The primary outcome is live birth rate, with secondary outcomes encompassing various pregnancy and perinatal parameters. Randomisation will be stratified by age and transfer type. Intention-to-treat analysis will be performed using generalised linear models. The trial will be monitored by an independent data and safety monitoring committee, including one interim analysis.

Ethics and dissemination

This study has been approved by the Institutional Ethics Committee of Northwest Women’s and Children’s Hospital (No. 2025-058-02). Written informed consent will be obtained from all participants. The study results will be disseminated at scientific conferences and published in peer-reviewed journals.

Trial registration number

NCT07185230.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• This study is a continuation and extension of previous research, designed as a rigorously conducted randomised controlled trial, providing robust evidence for the potential role of atosiban in improving assisted reproductive technology outcomes.

• The adoption of clearly defined selection criteria enables precise identification of high-risk populations most likely to benefit from atosiban intervention, thereby enhancing the clinical relevance and targeted applicability of the study findings.

• Although ultrasound assessment of uterine contractions is performed by two well-trained examiners in accordance with standardised protocols, this assessment method remains inherently operator-dependent.

• As it is anticipated that not all participants will complete the scheduled follow-up transvaginal ultrasound to reassess uterine contractile activity, the overall effect of atosiban on uterine contractions may not be comprehensively evaluated in the entire study cohort.

Introduction

Assisted reproductive technology (ART) is one of the most effective treatments for infertility. Since the birth of the world’s first child conceived by in vitro fertilisation (IVF) in 1978, more than 10 million infants have been born worldwide through ART, with numbers continuing to increase steadily.¹ Despite the increasing accessibility to ART and ongoing technological advancements, the overall success rate still faces a significant bottleneck: currently, the average clinical pregnancy rate per embryo transfer cycle is approximately 35%, and the live birth rate is only about 25%, with no significant improvement in recent years.² Optimising embryo transfer strategies to enhance implantation efficiency and improve live birth rates has thus emerged as a paramount focus in reproductive medicine.

Mounting evidence implicates uterine contraction as a critical determinant of implantation success. Excessive uterine contractions during the embryo transfer are thought to directly impair implantation by displacing embryos from the uterine cavity.3,6 A prior meta-analysis demonstrated a strong negative correlation between contraction frequency and clinical pregnancy rates: women with two or more uterine contractions at the time of the embryo transfer had a significantly lower clinical pregnancy rate than women with two or fewer contractions (OR 0.52, 95% CI 0.38 to 0.69, p<0.01).⁵ Emerging mechanistic studies further suggest that oxytocin, a key regulator of uterine motility, drives abnormal contraction patterns during this critical period. These findings have led investigators to hypothesise that targeting the oxytocin signalling pathway with pharmacological agents may mitigate contraction-related implantation failure.

Atosiban, a selective oxytocin receptor antagonist, inhibits uterine contractions through three complementary mechanisms: competitive blockade of oxytocin receptors, suppression of intracellular calcium release and inhibition of prostaglandin F2α synthesis. Despite its well-characterised mechanism of action, the clinical efficacy of atosiban in improving ART outcomes remains a topic of intense debate.⁷ Supportive evidence includes two large systematic reviews: one encompassing 11 randomised controlled trials (RCTs) reported a significantly higher clinical pregnancy rate (CPR) among atosiban recipients compared with controls (RR 1.31, 95% CI 1.13 to 1.51, I²=61%), while another meta-analysis of seven RCTs found a similar improvement in CPR (RR 1.49, 95% CI 1.18 to 1.88, I²=68%).^{8 9} In the subgroup of patients with recurrent implantation failure, atosiban intervention yielded even more benefits in CPR (RR 1.54, 95% CI 1.37 to 1.74, I²=0.0%).¹⁰ Conversely, a single-centre matched retrospective cohort study failed to demonstrate a significant benefit of atosiban use before frozen embryo transfer (FET).¹¹ This controversy stems from inherent limitations in existing studies, including small sample sizes, heterogeneous participant inclusion criteria, failure to distinguish between fresh and frozen embryo transfer cycles and methodological biases, all of which contribute to a high risk of confounding and limit the generalisability of findings. Therefore, a 2021 Cochrane Systematic Review echoed these concerns, concluding that the available evidence is of limited quality and low certainty, precluding definitive recommendations regarding the efficacy and safety of atosiban during the embryo transfer.¹²

To address these gaps, a large-scale, placebo-controlled, double-blind RCT was conducted to evaluate the role of atosiban in women with a history of implantation failure undergoing frozen-thawed blastocyst transfer (the Phase I Atosiban study).^{13 14} Results showed no significant differences in clinical pregnancy rate (treatment: 59.7% (328/549) vs control: 57.5% (317/551), RR 1.04, 95% CI 0.94 to 1.15) or live birth rate (treatment: 49.5% (272/549) vs control: 44.7% (246/551), RR 1.11, 95% CI 0.98 to 1.26) between the treatment and control groups.¹⁴ However, in a subgroup analysis of 163 participants (23% of total) with ultrasound-confirmed high-frequency uterine contractions (>4 times/min), atosiban administration revealed a promising trend towards improving live birth rate (51.9% vs 39.3%, absolute difference=12.6%, RR 1.32, 95% CI 0.94 to 1.86; p=0.11), though the small sample size prevented definitive conclusions.¹⁴

Given the adverse effects of abnormal uterine contractions on embryo implantation and reproductive outcomes, there is a compelling rationale to further explore the clinical utility of atosiban during the embryo transfer. Future research should prioritise two key areas: (1) validating uterine contraction frequency as a biomarker for treatment selection and (2) investigating the efficacy of atosiban in a well-defined cohort of women with implantation failure and abnormal uterine contractions. Such studies will not only provide rigorous evidence-based support for clinical decision-making but also enable the precise identification of the subset of women who are most likely to derive meaningful benefit from atosiban intervention.

Objective and hypothesis

The present study seeks to validate the following hypothesis: Among women aged 20–40 years with a history of at least one previous embryo implantation failure who are scheduled to undergo another single blastocyst-stage embryo transfer, if abnormal uterine contractions (>4 contractions per minute or a negative wave pattern at any frequency) are identified by transvaginal ultrasonography prior to the transfer, the administration of atosiban before the procedure will lead to a statistically significant improvement in live birth rate compared with placebo.

Methods and analysis

A total of 792 eligible participants will be enrolled and randomly allocated in a 1:1 ratio. Except for the randomised intervention, all other clinical procedures will be performed according to the standardised protocols of the research centre, including ovarian stimulation, fertilisation (IVF/intracytoplasmic sperm injection), endometrial preparation, the embryo transfer procedure and luteal phase support measures.^{15 16} A summary of the study design is presented in figure 1.

Figure 1. Flowchart of the study.

The study protocol has been approved by the Institutional Ethics Committee of Northwest Women’s and Children’s Hospital (approval No.: 2025–058-02) and registered on ClinicalTrials.gov (registration number: NCT07185230). The trial will be conducted at the Assisted Reproduction Centre of Northwest Women’s and Children’s Hospital in Xi'an, China, under the supervision of an independent Data and Safety Monitoring Committee (DSMC) and conducted in strict compliance with good clinical practice (GCP) and the principles of the Declaration of Helsinki (2002). Written informed consent will be obtained from all participants prior to study enrolment, and the full consent form employed in this study is included as online supplemental file 1. The first participant was enrolled on 6 December 2025, and the study is expected to be completed by 30 June 2027.

Participant eligibility

Eligibility criteria

Infertile women who fulfil all of the following inclusion criteria, based on the results and findings of the Phase I Atosiban study, are eligible for participation:

1. Scheduled for a single fresh or frozen-thawed blastocyst-stage embryo transfer, regardless of PGT status.

2. Have a history of one or more episodes of implantation failures in previous embryo transfer cycles.

3. Age between 20 and 40 years (inclusive).

4. Exhibit abnormal uterine contractions on transvaginal ultrasound performed on the morning of the scheduled blastocyst transfer, defined as either high-frequency patterns (>4 contractions per minute) or a negative wave pattern (FC direction) at any frequency.

5. Each woman could only participate in one transfer cycle.

Exclusion criteria

1. Immunologic or endocrine disorders, including thyroid dysfunction, antiphospholipid syndrome, hyperprolactinaemia and other severe systemic diseases (eg, hypertension, diabetes, etc).

2. Uterine malformations (eg, intrauterine adhesions, unicornuate, bicornuate or septate uterus).

3. Untreated hydrosalpinx or endometrial polyp.

4. Cycles that included in vitro maturation or oocyte donation.

5. Prior participation in the Phase I Atosiban study.¹⁴

6. Current participation in any other clinical trial.

7. Inability to provide informed consent to the study.

Sample size calculation

In the Phase I Atosiban study, among 163 participants with abnormal uterine contractions and a history of one implantation failure, the live birth rate was 51.9% (41/79) in the atosiban group compared with 39.3% (33/84) in the placebo group, resulting in an absolute difference of 12.6% (95% CI −2.6 to 27.8).¹⁴ For this Phase II trial, which includes women with more than one implantation failure, the live birth rate in the placebo group is conservatively assumed to be 35%. To detect a minimum clinically relevant advantage of 10 percentage points for atosiban, 752 participants (376 per group) are required (power 80%; alpha of 0.05). Accounting for a 5% dropout rate, the final planned sample size is set at 792 (396 per group).

Withdrawal procedure

Participants have the right to withdraw from the trial at any time during the process. The decision to withdraw will neither affect their conventional clinical treatments nor their relationship with clinicians. If consent is withdrawn prior to randomisation, the participant will be excluded, and none of their data will be analysed. After randomisation, participants may discontinue the study intervention or follow-up assessments. However, in line with ethical guidelines and data protection policies, their data will be retained for the primary intent-to-treat (ITT) analysis. A secondary per-protocol (PP) analysis will be performed on the subset of participants who completed the prescribed intervention.

Screening and informed consent

The attending physician will conduct initial eligibility screening based on inclusion criteria #1–3. For fresh embryo transfer cycles, this process begins on day 2 after oocyte retrieval; for frozen embryo transfer cycles, it starts on the day of progesterone supplementation initiation. Women deemed preliminarily eligible after this screening will receive detailed counselling from clinicians regarding the study specifics, including its parallel cohort design, the assessment of uterine contractions on the day of transfer and the subsequent evaluation against inclusion criterion #4 to determine randomisation eligibility (figure 1). Written informed consent will be obtained prior to the scheduled blastocyst transfer date. All participants who provide signed consent will be included in the cohort study, and those meeting criterion #4 will proceed to the randomisation stage (see below).

Uterine contraction measurement

Uterine contractions will be assessed on the morning of blastocyst transfer using the transvaginal ultrasound protocol validated in the Phase I Atosiban study.^{13 14} Key procedural steps include (1) gentle insertion of the probe to minimise cervical stimulation and (2) recording of waveform activity in the uterine sagittal plane for at least 5 min. The observed waveforms are classified into the following categories: cervix-to-fundus (CF), fundus-to-cervix (FC), opposite-polarity, random or no activity. Abnormal contractions are defined as high-frequency patterns (>4 contractions per minute) or a negative wave pattern (FC direction) at any frequency.

Randomisation, allocation concealment and blinding

After the uterine contraction assessment is completed, participants will be divided into two independent pathways based on the results: individuals with abnormal uterine contractions will be randomised in a 1:1 ratio to receive either intravenous atosiban or a placebo (0.9% normal saline). Those demonstrating normal uterine contractions will not be randomised but will be included in a parallel observational cohort, where they will continue to receive standard clinical care. This cohort is designed to independently evaluate the association between uterine contraction status and key reproductive outcomes (implantation and pregnancy rates), irrespective of the atosiban intervention. Data from this cohort will be analysed and reported separately from the Phase II RCT (see figure 1).

Randomisation will be executed using a secure web-based clinical trial management system (Research Electronic Data Capture (REDCap)). An independent statistician will generate the allocation sequence via computer-driven blocked randomisation with varying block sizes of 4 or 6, stratified by female age (<35 years vs ≥35 years) and embryo transfer type (fresh vs frozen). The study coordinator will log in to REDCap to initiate randomisation, and the allocation outcome will be disclosed solely to the research nurse, who is not involved in patient care or outcome assessment. This nurse will prepare the study drugs by dispensing either atosiban or a placebo (0.9% normal saline) into intravenous bags that are identical in appearance. Uniform packaging and labelling will be employed to ensure no distinguishable features exist between the two groups. Blinding will be implemented across all study participants, clinicians (including the embryo transfer physician), laboratory personnel, outcome assessors and data analysts. The blind will be rigorously maintained throughout the entire study period until the conclusive statistical analysis is finalised. Unblinding will be permitted only for urgent clinical reasons and must strictly adhere to the formal protocol.

Treatment procedure

Approximately 30 min prior to embryo transfer, participants randomised to the atosiban group will receive an intravenous administration of atosiban (Tractocile; Ferring Pharmaceuticals, Geneva, Switzerland), starting with a bolus injection of 6.75 mg/0.9 mL delivered over 1 min followed by a continuous infusion: initially at 18 mg/hour for 1 hour, then reduced to 6 mg/hour for another 2 hours, culminating in a total administered dose of 37.5 mg.¹⁴ Participants allocated to the placebo group will undergo an identical administration schedule using 0.9% normal saline, which is packaged and labelled identically to the atosiban. One hour after embryo transfer, all randomised participants will undergo a follow-up transvaginal ultrasound to re-evaluate uterine waveform activity.¹⁷ To ensure consistency and minimise interobserver variability, this reassessment will be performed by the same operator who carried out the pretransfer evaluation.

Parallel observational cohort

Participants undergoing embryo transfer without abnormal uterine contractions will be ineligible for the atosiban-placebo RCT. Instead, they will be enrolled in a parallel observational cohort and proceed directly to single blastocyst transfer in accordance with the centre’s standard clinical protocol. The reproductive outcomes of this cohort will be compared against those of the randomised cohort (comprising participants with abnormal contractions, ie, the combined atosiban and placebo groups). This comparative analysis is designed to evaluate the impact of abnormal uterine contractions on reproductive outcomes and to determine whether Atosiban can mitigate its potential adverse effects.

Outcome measures

Primary outcomes

The primary outcome is the live birth rate following a single embryo transfer. Live birth is defined as the delivery of at least one newborn at ≥24 weeks of gestation.¹⁸ For twin pregnancies, if at least one infant survives, it is counted as one live birth for the mother.¹⁹

Secondary outcomes

Secondary outcomes include additional pregnancy and obstetric events, reported in accordance with the international consensus standards set by the core outcome set for infertility research.¹⁹ Definitions are specified below.

1. Pregnancy outcomes

   • Biochemical pregnancy: defined as serum beta-human chorionic gonadotropin (β-hCG) >20 IU/L 12 days after embryo transfer.

   • Clinical pregnancy: defined as the presence of at least one gestational sac confirmed by ultrasonography (ectopic pregnancy is included as a clinical pregnancy).

   • Ongoing pregnancy: defined as the presence of a gestational sac and fetal heartbeat after 12 weeks of gestation.

   • Miscarriage: defined as the spontaneous loss of an intrauterine pregnancy before 20 completed weeks of gestation.

   • Ectopic pregnancy: defined as a pregnancy implanted outside the uterine cavity and diagnosed by ultrasonography, surgical visualisation or histopathology.

   • Multiple pregnancy: defined as a pregnancy with two or more gestational sacs or fetal heartbeats at 7 weeks of gestation.

2. Obstetric and perinatal outcomes.

For women who achieve an ongoing pregnancy, comprehensive maternal and perinatal outcomes will be prospectively recorded. These include hypertensive disorders of pregnancy (encompassing chronic hypertension, gestational hypertension, pre-eclampsia and eclampsia), placental implantation abnormalities (placenta previa, placenta accreta spectrum, placental abruption and/or abnormal cord insertion), preterm birth, preterm premature rupture of membranes, mode of delivery, gestational age at delivery, birth weight (categorised as very low birth weight, low birth weight, average birth weight and macrosomia), size for gestational age (small for gestational age and large for gestational age), admission to the neonatal intensive care unit, stillbirth or late fetal death, early neonatal death (within 28 days of birth) and congenital malformations or birth defects.

Uterine contraction outcomes

The uterine contraction parameters measured before and after embryo transfer will be fully documented and retained, including contraction frequency and direction (eg, frequency of CF contractions per minute and frequency of FC contractions). Furthermore, participants’ contraction experiences before and after the transfer will also be documented as part of the exploratory analysis (online supplemental file 2).

Data collection and management

All baseline characteristics, clinical outcomes and laboratory results will be routinely documented in the participants’ electronic health records. For the enrolled participants, these data will be extracted by researchers at the IVF centre using customised electronic case report forms and entered into a dedicated electronic database. Rigorous validation checks will be implemented to ensure the accuracy and completeness of the captured information.

Statistical analysis

Baseline characteristics will be summarised using descriptive statistics, and intergroup balance will be assessed visually. Continuous variables will first undergo normality testing via frequency histograms and the Shapiro-Wilk test; non-normally distributed data will be reported as median and IQR. Categorical variables will be presented as counts and proportions for each group. Additional reporting will include the number of participants recruited, those lost to follow-up, protocol deviations and other relevant descriptive data. The ITT set will include all randomised women and serve as the primary dataset for all key comparisons between groups. This approach strictly adheres to the intention-to-treat principle to preserve the benefits of randomisation-induced baseline balance. The PP set will consist solely of women who adhered strictly to the trial protocol and will be used for secondary analyses. Primary outcome (live birth) and other binary outcomes will be performed using generalised linear models with a binomial distribution and log link function (ie, log-binomial models). If the log-binomial model fails to converge, a modified Poisson regression model with robust SEs will be employed as a substitute. All intergroup outcome comparisons will be adjusted for the two stratification factors (maternal age and type of embryo transfer) to control potential confounding between treatment assignment and stratification variables.

Regarding missing observations, pregnancy-related endpoints (β-hCG, clinical pregnancy and ongoing pregnancy) will be imputed based on available data from subsequent pregnancy assessments (eg, a missing β-hCG result will be imputed as ‘positive’ if clinical pregnancy is subsequently confirmed). Endpoints not verifiable through subsequent assessment (eg, live birth) will be treated as missing values. Adverse event documentation will apply worst-case-scenario imputation exclusively to causality, severity and intensity determinations. Detailed statistical procedures will be fully specified in a standalone statistical analysis plan (SAP). This SAP will be finalised prior to database lock to ensure full transparency and reproducibility of the entire analytical process.

Prespecified subgroup analysis

Prespecified subgroups will be established along the following dimensions:

1. Fresh versus FET cycles.

2. Two types of abnormal contractions: solely high-frequency peristalsis (>4 times/min) versus negative waves (fundus to cervix) of any frequency.

3. Frequencies of contractions (dichotomising the cohort at the median value of abnormal contractions per minute).

4. Women aged <35 years versus ≥35 years.

5. PGT-A versus non-PGT-A blastocysts.

6. High-quality (≥BB) versus fair-quality blastocysts (BC, CB and CC).²⁰

To explore potential effect modification, we will model the interaction between treatment allocation and the subgrouping variable, supplementing this with subgroup-specific analyses of the primary outcome.

Data safety monitoring committee

An independent DSMC will oversee trial safety and integrity, comprising one independent chair and three expert members specialising in ART clinical practice, embryology and biostatistics. The DSMC will conduct quarterly reviews of trial progress and provide recommendations to the trial steering committee (TSC). A single interim analysis will be triggered upon reaching 50% of planned enrolment. This analysis will be executed by an independent statistician who will provide only aggregated results to the DSMC for their evaluation. Based on this interim safety and efficacy assessment (focusing particularly on ongoing pregnancy rates), the DSMC may issue guidance concerning trial continuation. Throughout this process, blinding will be rigorously maintained. The TSC will remain blinded to interim results, and all investigators and research staff will maintain therapeutic allocation concealment. To preserve the overall two-sided type I error rate at 0.05, the O'Brien-Fleming alpha-spending function will be applied, ensuring the final analysis uses conventional thresholds. Unblinding will not be considered unless there are substantial safety or trial integrity concerns that require investigation. Following DSMC recommendations, the TSC holds ultimate responsibility for decisions regarding trial termination based on efficacy or safety grounds.

Follow-up procedure

All women with confirmed clinical pregnancy will continue to be followed by the research nurses to document pregnancy events and outcomes at 12 and 24 weeks of gestation, completion of pregnancy and approximately 6 weeks postpartum. Participants with negative pregnancy tests will not require further follow-up within this trial. Loss to follow-up will be determined if repeated contact attempts at various intervals yield no response. According to the mandatory reporting requirements issued by the Chinese Society of Reproductive Medicine, all women achieving pregnancy via ART in China will undergo sustained monitoring by their treating IVF centres. The relevant data will typically be collected by clinical staff via telephone communication with the couples.

Safety and adverse event reporting

The investigator will immediately inform participants and the accredited Medical Research Ethics Committee (MREC) if anything occurs on the basis of which it appears that the disadvantages of participation may be significantly greater than was foreseen in the research proposal. Unless suspension would directly endanger participant health, the study will be suspended pending further review by the MREC. Investigators will take care that all participants are kept informed. All adverse events (AEs), whether observed or spontaneously reported by participants and regardless of treatment group or suspected relationship to the intervention, will be recorded. An AE is defined as any undesirable medical occurrence during the trial, irrespective of causality. Serious AEs (SAEs) refer to an event meeting any of the following criteria: death (maternal, fetal or neonatal), life-threatening situations, hospitalisation or prolongation of existing hospitalisation, permanent or severe disability, congenital anomalies or birth defects or newly identified safety concerns potentially impacting participant well-being (eg, ovarian hyperstimulation syndrome, ectopic pregnancy, miscarriage and pre-eclampsia). All SAEs will be reported to the MERC in accordance with its requirements.

Declarations

Ethics and dissemination

The study protocol has obtained ethical approval from the Ethics Committee of Northwest Women’s and Children’s Hospital (approval no. 2025–05802) and was also prospectively registered on ClinicalTrials.gov under identifier NCT07185230. The study will be conducted in strict compliance with GCP guidelines and the principles of the Declaration of Helsinki. An independent DSMC will be established to oversee trial conduct, protect participant safety and monitor data integrity throughout its duration. Written informed consent, which clearly outlines the associated rights and potential risks, will be obtained from all participants. The study results will be submitted for publication in peer-reviewed open-access journals and presented at relevant national and international scientific conferences.

Patient and public involvement

Prior to the finalisation of the protocol, women scheduled to undergo ART with single blastocyst transfer were invited to participate in the research patient panel. These participants offered valuable insights regarding intervention burden, personal experiences and preferences and were also given the opportunity to provide feedback on the study design. In the future, they will contribute to the dissemination of the study results but will not participate in the study’s recruitment process or implementation. Additionally, members of the public will not be involved in the study’s design, recruitment, execution or interpretation of results. Prior to publication, we will discuss the study results within the research patient panel.