Background

Pre-eclampsia (PE), which affects about 5% of Chinese pregnant women,1 is one of the main causes of adverse maternal, fetal and neonatal outcomes, especially when the disorder is early onset requiring delivery before 37 weeks’ gestation (preterm-PE).2–5 Extensive research has been devoted to establishing an effective screening model based on maternal characteristics, biophysical and biochemical parameters, and identifying prophylactic interventions for preterm-PE.6

Prediction of pre-eclampsia

PE has been found to be associated with altered angiogenesis, which involves an imbalance of proangiogenic factors such as vascular endothelial growth factor and placental growth factor (PlGF) and antiangiogenic circulatory factors such as soluble fms-like tyrosine kinase-1 (sFlt-1) and soluble endoglin.7 Changes in the levels of these factors are linked to the clinical manifestation of PE.8 Therefore, these angiogenic factors have the potential to serve as biochemical markers for predicting the onset of PE and related adverse outcomes.9 Among these factors, PlGF, which has both vasculogenic and angiogenic functions, has been demonstrated to be the most promising biomarker for predicting PE in the first trimester.10 In addition, uterine artery pulsatility index (UtA-PI) and mean arterial pressure (MAP) have also been identified to be effective predictors for PE.11 A recent study has established a new approach for early screening for PE by using the Bayes’ theorem to combine the prior information from maternal demographic characteristics, medical and obstetrical history with the aforementioned biomarker multiples of the median values (known as Fetal Medicine Foundation (FMF) triple test), which is based on a survival time model for the time of delivery for PE. The study involving nearly 60 000 singleton pregnancies has demonstrated that the first trimester triple test achieves an estimated detection rate of 75% for preterm-PE at a 10% false positive rate (FPR).12 Another prospective multicentre study implemented in England demonstrates that the performance of first-trimester screening for PE by a combination of maternal factors and biomarkers is superior to that achieved by the method recommended by the National Institute for Health and Care Excellence (NICE) guidelines.13 14 Our recent prospective study, which has recruited more than 10 000 Asian women including approximately 8000 Chinese women, has proven that the Bayes’ theorem-based method for the first trimester screening of preterm-PE performs better than the screening method recommended by the NICE.15

Aspirin and pre-eclampsia

The role of aspirin in the prevention of PE has been evaluated by numerous randomised trials,16–18 yet there remains a question about the best treatment regime. An individual patient data meta-analysis has reported that the administration of low-dose aspirin in high-risk pregnancies is associated with a decrease in the rate of PE by approximately 10%.19 A prospective multicentre randomised controlled trial (RCT)—Combined Multi-marker Screening and Randomised Patient Treatment with Aspirin for Evidence-Based pre-eclampsia Prevention (ASPRE)—has demonstrated that treatment with low-dose aspirin (150 mg daily) versus placebo in women at high-risk for preterm-PE identified by the algorithm combining maternal history and characteristics, biophysical (MAP and UtA-PI) and biochemical markers (pregnancy-associated plasma protein A and PlGF) results in a lower incidence of preterm-PE (OR 0.38; 95% CI: 0.20 to 0.74; p=0.004).16 20 In the most recent RCT conducted in Mainland China, including 1105 high-risk cases randomised into placebo (n=284) and aspirin groups (including three subgroups: 272 cases in aspirin 25 mg group, 278 cases in aspirin 50 mg group and 271 cases in aspirin 75 mg group), and the results showed that low-dose aspirin significantly reduced the incidence of early-onset PE and that such prevention was associated with a dose-response effect with the reduction rates of 49%, 25.8% and 13.8%, respectively, in each aspirin group compared with placebo.21

Metformin and pre-eclampsia

Metformin has been used extensively in the treatment of gestational diabetes and polycystic ovary syndrome, and no adverse fetal effects have been reported in the first, second and third trimesters of pregnancy.22–24 Metformin can cross the placenta freely, while being poorly metabolised by the fetus, as many studies have shown no evidence of a higher chance of premature labour, macrosomia and morbidity or mortality of the child up to first year of life.25 However, children exposed to maternal intake of metformin during pregnancy have been shown to have a higher body mass index (BMI) later in life.25

In recent years, metformin has been proposed for the prevention of PE because it has been shown to reduce sFlt-1 and soluble endoglin secretion from endothelial cells and primary trophoblasts.26 An RCT of metformin versus placebo for the prevention of fetal macrosomia in women with BMI >35 kg/m2 without diabetes demonstrated a lower incidence of PE in the metformin group (n=292) than in the placebo group (n=198) (3.0% vs 11.3%; p=0.001).27

Another recent trial of metformin versus placebo given to women with preterm-PE indicated that in the metformin group participants delivered their newborns at a median of 17.5 days after randomisation, 9.6 days longer than the 7.9 days randomisation-to-delivery interval of women in the placebo group and their babies had a shorter stay in the neonatal nursery.28 A meta-analysis of 15 studies including 3124 pregnancies demonstrated that the posterior probabilities of metformin, at a median dosage of 1.5 g, having a beneficial effect on the prevention of PE, pregnancy-induced hypertension and any hypertensive disorders of pregnancy (HDP) were 92.7%, 92.8% and 99.2%, respectively, when compared with any other treatment or placebo. These findings suggest that metformin use has a high probability of being associated with reduced incidence of HDP.29 Another meta-analysis of 23 studies including 6301 pregnancies concluded that there was a significant reduction in the likelihood of PE in women randomised to metformin compared with any other treatment (OR 0.69, 95% CI: 0.50 to 0.95; I2=55%, p=0.02).30 Given the potential benefits of metformin in pregnant women who are at high-risk of PE, we aim to explore whether the administration of 1.5 g metformin with aspirin, compared with aspirin alone, is more efficacious in reducing the incidence and severity of PE among high-risk pregnant women.

Current recommendation for screening and prevention of pre-eclampsia

The current guidelines in Mainland China recommend that women should be risk stratified for PE based on a checklist of maternal demographic characteristics, medical and obstetrical history, which is similar to those recommended by the American College of Obstetricians and Gynaecologists (ACOG) and the UK NICE guidelines.31 The performance of the screening approach recommended by ACOG and NICE is however suboptimal, which achieves detection rates of 54.6% (95% CI: 43.5% to 65.9%) and 26.3% (95% CI: 16.3% to 36.3%), respectively, at FPRs of 20.4% and 5.5%, in a large Asian population.15

The recommended strategy for preventing PE is to treat with aspirin prophylaxis at 50–150 mg daily from 12 to 16 weeks’ gestation until 26–28 weeks’ gestation.31 There is no consensus about the preferred aspirin dosage for the prevention of preterm-PE. In routine clinical practice, aspirin prophylaxis at 75 mg daily is currently the mainstream dosage. However, extensive studies have demonstrated that aspirin is effective in reducing the incidence of preterm-PE only when it is initiated before 16 weeks’ gestation and at a daily dose of 100 mg and above.17 Furthermore, the ASPRE trial has demonstrated that aspirin 150 mg can reduce the rate of preterm-PE by 62% compared with placebo.16 Key international professional bodies such as the International Federation of Obstetrics and Gynaecology, the International Society for the Study of Hypertension in Pregnancy (ISSHP) and the International Society of Ultrasound in Obstetrics and Gynaecology, have recommended aspirin 150 mg daily for clinical practice worldwide.32–34 However, so far, no previous study has compared the preventive ability of aspirin 150 mg daily versus aspirin 75 mg daily with or without metformin 1.5 g daily in a large scale of the Chinese population. A prospective multicentre RCT in Mainland China is therefore urgently needed for providing high-quality evidence to support the implementation of aspirin prophylaxis at the optimal dosage to prevent preterm-PE. Hence, in this trial, we will evaluate whether a dosage of 150 mg daily is superior to 75 mg daily with or without metformin 1.5 g daily.

AVERT trial

In the AVERT trial, effective preterm-PE screening will be performed at 11–13 weeks’ gestation to enable the maximum benefit of aspirin and metformin among high-risk women.

Hypothesis

We hypothesise that aspirin 150 mg daily and aspirin 75 mg with metformin 1.5 g daily, compared with aspirin 75 mg daily, commenced from the first trimester of pregnancy in Chinese women at increased risk for preterm-PE will be more efficacious in reducing the incidence and severity of the disease.

Methods

The AVERT trial will be a multicentre, double-blind, 3-arm RCT, which will be conducted from May 2023 to November 2025. All eligible women attending for the 11–13 weeks visit will be invited to have PE screening by the combination of maternal factors, MAP and PlGF measurements. The decision to implement the screening test without UtA-PI is based on the challenges that antenatal ultrasound scans are routinely performed in the Radiology Department in Mainland China and not all sonographers are holders of the FMF uterine artery Doppler measurement competency certificate. Written informed consent will be obtained from eligible women who agree to undergo screening. Maternal MAP will be measured according to the standardised protocol.37 Maternal serum PlGF will be measured using automated machines (Superflex system, Revvity Inc. former PerkinElmer Life and Analytical Sciences, Shanghai, China; cobas e411 system, cobas e601 system, Roche Diagnostics, Rotkreuz, Switzerland; iMAGIN 1800, Ningbo Aucheer Biotechnology, Ningbo, China) that provide reproducible results. Following the screening, high-risk women, based on a risk cut-off that corresponds to a screen-positive rate of 10% for preterm-PE, will be invited to take part in the RCT by the designated members of the trial teams (figure 1). Women eligible to participate in this trial will receive written information on the test drugs and provide informed consent. Each consented participant will be assigned a screening number and a randomisation code. The randomisation code will determine who receives aspirin 75 mg/day, aspirin 150 mg/day or aspirin 75 mg/day with metformin 1.5 g/day, from 11 to 13 weeks. The independent statistician will keep the randomisation code list. All participants, the principal investigator, participating research doctors and clinical trial pharmacists will remain blind to trial drug allocation.

Figure 1

Flow diagram of participants in the screening study and the randomised controlled trial. CRL, crown rump length; IMP, investigational medicinal product; MAP, mean arterial pressure; PlGF, placental growth factor; wk, week.

Data collection

Participant data for this study will be entered into an electronic case report form (eCRF) and signed by the enrolling researcher. Data on pregnancy outcomes will be collected from the hospital maternity records. Trial participants will complete their drug daily and record any side effect on the eClinicaltrial Patient Management (eCPM) application programme.

Randomisation

An independent statistician will generate the randomisation sequence using a permuted block in multiples of 3, stratified according to participating centres. When a participant enrols in the study, the Interactive Response Technology (IRT) system will generate a randomisation code, which corresponds to a treatment pack with the same code at a given participating centre. Each treatment pack containing cartons of bottles will only be identified by a randomisation code. The treatment allocation will only be revealed to the researchers after completion of the study or where clinically essential.

Concealment of allocation

If an investigational medicinal product (IMP) is being dispensed from the pharmacy, the participant or a member of the clinical team on behalf of the participant, will bring the request form with the allocated randomisation code to the pharmacy, where the pharmacy will provide the IMP treatment pack matched to the randomisation code. The participants will be blinded to the IMP allocation. All matching capsules will be identical, and therefore it will not be possible to distinguish the capsules with active medicine and the ones with matching placebo.

Frontage Laboratories, Co., Ltd, Suzhou, China, will provide IMP labelling (for all treatment packs, cartons and bottles). Wuxi Tigermed Consulting Co.,Ltd, Jiangsu, China, will keep the randomisation code list confidential to maintain the blind, and the randomisation code list will be transferred to the IRT system to enable the online randomisation and establish emergency unblinding service.

Intervention

A total of eight capsules per day will be prescribed to the high-risk women, including six capsules of metformin or matching placebo and two capsules of aspirin or matching placebo. Aspirin (or an identical-appearing placebo) and metformin (or an identical-appearing placebo) are in different forms of medication and provided in different bottles, making them easy to be distinguished.

Randomised participants will be advised to start the trial drugs within 24 hours of randomisation as presented in online supplemental table 2.

Laboratory tests

At the time of the 11–13 weeks visit, 10 mL of maternal blood will be taken for the measurement of PlGF as described above. Where feasible, the remaining serum will be stored at −80°C for future studies of potential biochemical markers for pregnancy complications.

Participant compliance

At each follow-up visit, the drug diary in eCPM and drug compliance will be reviewed by the trial team. Research staff at each participating centre will also ask about drug compliance at telephone/WeChat follow-up. Participants will be encouraged to report any concerns or side effects in the drug diary for review at each follow-up visit. The following definitions will be used to assess compliance in terms of check counts in the drug diary38: very good compliance (≥90%), good compliance (<90% and ≥80%), moderate compliance (<80% and ≥50%) and poor compliance (<50%).

Patient and public involvement

Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

OutcomesPrimary outcome

PE is defined according to the guidelines of the ISSHP33:

Secondary outcome

As defined above in the Secondary objectives section.

Collection of pregnancy and neonatal outcomes

Data on pregnancy and neonatal outcomes will be collected from the hospital maternity records or their general medical practitioners. Data will be entered into eCRF. The obstetrical records of the randomised women with pre-existing or pregnancy-associated hypertension will be examined to determine if the condition was chronic hypertension, PE or gestational hypertension. In the event neonates are admitted to Special Care Baby Unit (SCBU), additional neonatal outcomes will be collected from the discharge summary of SCBU.

Side effects and adverse events reporting

Adverse event (AE) and reaction will be recorded for high-risk women participating in the RCT. Safety evaluations will be conducted at each of their follow-up visits. The period for AE reporting for events directly related to the participant will be from the time of the first dose until 30 days post final IMP administration. The period for AE reporting for congenital abnormalities or birth defects identified in participant’s baby will be 30 days after delivery or 30 days after the estimated due date, whichever is the latter. The participants will be followed up by a telephone/WeChat interview 30 days after the last dose of IMP. All events will be followed until resolution if that means beyond the timelines defined here.

If the event is classified as non-serious, then it should be recorded in the patient’s medical notes and entered into eCRF. If the event is classified as serious then a serious AE (SAE) form must be completed and should be reported to the sponsor notified within 24 hours of the investigator becoming aware of the event.

The following events are defined as protocol-defined exceptions to SAE-reporting and should only be reported to the sponsor within 7 days of the investigator becoming aware of an event as an SAE if the investigator believes the event is a result of the AVERT intervention: hospitalisation for maternal or fetal observation, preterm delivery, miscarriage, stillbirth or neonatal death, admission of the baby to neonatal intensive care unit, termination for fetal or maternal indication. Admission for delivery and Caesarean section is an exception to SAE reporting which does not require reporting if it is thought to be part of the routine progress of the pregnancy.

Statistical analysis planSample size

When using 4.3% as the incidence of preterm-PE of base-case,16 aspirin 75 mg and aspirin 150 mg are expected to reduce the incidence to 3.9%19 and 1.6%, respectively.16 19 A sample size of 790 per group is required to detect a significant difference with 80% power and type I error of 5%. In a 3-arm design, after adjusting the inflation of type I error using Bonferroni correction, a sample size of 957 per group is required so in total 3000 high-risk women are required to be randomised. Assuming a 70% uptake, 4285 high-risk women are required to be identified. The detection rate of the first trimester combined test of maternal factors, MAP and PlGF is 65% and we expect a 10% screen-positive rate, therefore, 66 000 women with singleton pregnancy will need to be screened.

Analysis population

The data analysis will be primarily performed according to the intention-to-treat (ITT) principle, that the ITT population will include all the subjects who have been randomised in this study. An additional per-protocol analysis will be conducted on the per-protocol populations, defined as the study population having (≥90%), good compliance (<90% and ≥80%) of the trial drug.

Descriptive Statistics

A full set of descriptive statistics for all variables, overall and by treatment group, will be produced. Graphical displays will be produced as appropriate.

Primary analysis

The primary analysis will be a mixed-effects logistic regression analysis of the incidence of preterm-PE adjusted with the fixed effects of treatment, baseline characteristics, estimated risks and random effects for recruiting centre. A choice of transformation of risk (eg, logit transform) or grouping into levels will be made based on a blinded review of the data. The treatment effect will be tested at the two-sided 5% level. 95% CIs will be produced for the proportions developing preterm-PE in each of the study arms.

For planned secondary analysis of the primary outcome, survival analysis of the time to delivery with PE will be conducted, and births for other causes will be treated as censoring. Pre-specified baseline variables considered to be predictive will be included as appropriate. We will investigate the interactions of both the abovementioned variables and gestational age at randomisation with the treatment effect. This analysis of treatment interactions will be considered as exploratory.

Secondary analyses

Similar to primary analysis, a mixed effects logistic regression analysis will be conducted to compare the secondary outcomes. To compare the gestational age at delivery between treatment groups, a multivariate linear regression analysis with a similar adjustment will be conducted.

Safety analysis

The incidence rates of AEs and SAEs and their relationship to trial drugs will be summarised by treatment group. The proportion of women discontinuing treatment will be summarised by reason and by treatment group.

Subgroup analysis

Subgroup analysis will be conducted to compare the primary and secondary outcomes by age groups, BMI and obstetrical history. The details of subgroup analyses will be noted in the statistical analysis plan.

Trial oversight

The Trial Management Team will be set up to assist with the co-ordination and day-to-day operational issues in the management of the trial, including budget management and strategic management of the trial. The Independent Data and Safety Monitoring Committee (IDSMC) is independent of the trial and is the only oversight body that has access to unblinded accumulating comparative data. The IDSMC is responsible for monitoring the progress of the trial including: recruitment, protocol adherence, AEs and side effects of treatment as well as the difference between the trial treatments on the primary outcome measures. The IDSMC will be appointed and will meet at least annually to review the safety data. They will provide a confidential trial progress report at the end of each meeting, which will be sent to the Trial Steering Committee (TSC). The TSC is the independent group responsible for oversight of the trial in order to safeguard the interests of trial participants. The Chinese University of Hong Kong Shenzhen Research Institute is the trial sponsor and is responsible for securing the arrangements to initiate, manage and finance the trial.

Ethics approval and dissemination

The trial will be conducted in compliance with the principles of the Declaration of Helsinki (2013), the principles of Good Clinical Practice and in accordance with all applicable regulatory requirements including the state health and Family Planning Commission of the People’s Republic of China measures for ethical review of biomedical research involving human beings (涉及人的生物医学研究伦理审查办法, 2016, No.11, https://www.gov.cn/gongbao/content/2017/content_5227817.htm) and the State Drug Administration on the quality management of drug clinical trials (临床试验质量管理规范, 2020, No.57, http://www.gov.cn/zhengce/zhengceku/2020-04/28/content_5507145.htm).

A favourable ethical opinion was obtained from Joint Chinese University of Hong Kong-New Territories East Cluster Clinical Research Ethics Committee (Reference number 2021.406). Subsequent approval by individual ethics committees and competent authority was granted. It is intended that the results of the study will be reported and disseminated at international conferences and in international peer-reviewed scientific journals.

On completion of the trial, and after publication of the primary manuscript, the individual participant data will not be shared according to the Data Security Law and the Personal Information Protection Law of the government of People’s Republic of China. The summary data could be shared.

The trial is registered with ClinicalTrials.gov.