Candidates selection for oocyte donation in a public gamete bank—Predictive value of the anti-Müllerian hormone

Abstract

Background:

Infertility treatments with oocyte donation are becoming frequent. Recruitment of oocyte donors is a demanding and costly process and therefore of crucial importance. The selection of the oocyte donors undergoes a rigorous evaluation process of the candidates with routine measurement of the anti-Müllerian hormone (AMH) levels (ovarian reserve test). Our aim was to assess whether AMH levels could act as a good marker as tool to select the donor candidates and correlate them with the ovarian response to stimulation with a gonadotropin-releasing hormone antagonist protocol as well as to identify and validate the appropriate AMH level threshold by correlating it with the number of oocytes retrieved.

Methods:

A retrospective analysis of the oocyte donors' clinical records was performed.

Results:

The mean age of the participants was 27 years. The ovarian reserve evaluation showed a mean AMH of 5.20 ng/mL. An average number of 16 oocytes was retrieved (12 mature oocytes MII). AMH levels showed a statistically significant positive correlation with the number of total oocytes retrieved. A threshold value of AMH = 3.2 ng/mL predictive of the retrieval <12 oocytes (areas under the curve, 0.7364; 95% confidence interval: 0.529–0.944) was identified by receiver operating characteristic curve. Using this cutoff, the normal response (12 oocytes) was predicted with a sensitivity of 77% and a specificity of 60%.

Conclusions:

The measurement of AMH may be a determining factor in the choice of the oocyte donor candidates to maximize the response to requests from beneficiaries who require donor oocytes to perform assisted reproductive technique cycles.

Introduction

Infertility treatments with oocyte donation are becoming more frequent.¹ Recruitment of oocyte donors is a demanding and costly process, and many of the candidates who initially show willingness to join the program end up giving up and approximately 20% fail the rigorous selection process for medical or psychological reasons.^1,2 The recruitment of donors is therefore of crucial importance. All the evaluation tools of the candidates, namely the ovarian reserve tests, are fundamental in the selection of the candidates.¹

Characteristics of donors, such as age or body mass index (BMI), and ovarian reserve tests, such as baseline follicle-stimulating hormone assay or antral follicle count (AFC), have been used as predictors of ovarian response in oocyte donation cycles.^1,3 Only age and the AFC proved to be effective.³ More recently, the role of anti-Müllerian hormone (AMH)—already known in patients with infertility⁴—has been validated as a predictor of ovarian response in cycles of oocyte donor stimulation.^1,5 AMH circulating levels, which is produced by the granulosa cells of the small antral follicles, reflect the primordial follicle pool size and are considered to be the best biochemical surrogate marker of the ovarian function. AMH levels seem to be an early, reliable and (in)direct indicator of declining ovarian function, very useful in the clinical setting when evaluating fertility in women.^4,6 Hamdine et al (2015) demonstrated that AMH is an adequate predictor for both high and low response in gonadotropin-releasing hormone (GnRH) antagonist cycles for assisted reproductive techniques (ART) and the optimization and individualization of this protocols may be improved by using an AMH-tailored approach.⁷ A systematic review and meta-analysis performed in 2022 revealed that AMH had better prediction for live birth in advanced age women and may have implicative predictive value for assisted conception counseling of couples of advanced ages when performing ART.⁸ The potential use of AMH measurement in selection of oocyte donor candidates has been asserted by some authors.^1,5 Martínez et al (2013) conducted a study at the Institut Universitari Dexeus in Barcelona to establish the reference values of AMH in their population of oocyte donors, correlating it with the response obtained in an ovarian stimulation protocol with GnRH antagonists, to determine the capacity of AMH in the prediction of the poor responders.⁵ In this study, the AMH cutoff to predict a poor responder candidate was 2.31 mg/mL.⁵

The Portuguese public gamete bank (PGB) was established in 2011 at Centro Materno-Infantil do Norte Dr Albino Aroso—Centro Hospitalar Universitário do Porto in Oporto. The selection of the oocyte donors undergoes a rigorous evaluation process of the candidates, consisting of a fertility assessment consultation, a psychology consultation, conducting blood tests, genetic study (with prior consultation in Medical Genetics), and conducting tests of evaluation of the ovarian reserve. Routine measurement of the AMH levels was routinely implemented to every Portuguese public bank oocyte donor candidates since 2017. Our aim was to assess whether AMH levels could act as a good marker as tool to select the donor candidates and correlate them with the ovarian response to stimulation with a GnRH antagonist protocol and to identify and validate the appropriate AMH level threshold by correlating it with the number of oocytes retrieved.

Methods

Subjects and study design

Retrospective analysis of the oocyte donors' clinical records from the PGB (Oporto). This study was performed at the Centro Materno Infantil do Norte—Centro Hospitalar Universitário do Porto with data collection from oocyte donors recruited between January 2017 and August 2018 (only oocyte donors with the AMH level determined). The principles of the Declaration of Helsinki were followed. The approval by the Institutional Ethics Committee was obtained. The inclusion criteria were women between 18 and 34 years plus 364 days of age wishing to undergo oocyte donation, who fulfilled all the clinical (normal menstrual cycles between 25 and 35 days, lack of relevant personal pathology and inherited family pathology, BMI between 18.5 and 30 kg/m², and AMH level > 1.2 ng/mL), and legal requirements (according to Portuguese law, until June 2021, each woman could only make three oocyte donations, at intervals of at least six months, after that date, each candidate can only perform four oocyte donations at intervals of at least 3 months, and from April 24, 2018, only candidates for gametes donation who accept the nonanonymity of the process can be recruited). All candidates to oocyte donation fulfilling the inclusion criteria underwent a physical examination, full blood test, prolactin and thyroid function measurement, serologies such as for syphilis, viral markers (HIV, hepatitis B and C infection, and other virus according to candidates travel history such as CMV and HTLV), karyotype determination, research of autosomal recessive diseases more prevalent in the Portuguese population (cystic fibrosis, spinal muscular atrophy, Fragile X syndrome, and hemoglobinopathies),⁹ ultrasound scan, and AMH measurement (irrespective of the day of the menstrual cycle).

All donors were pretreated with contraceptive pills and received a GnRH antagonist protocol. Ovarian stimulation was performed with recombinant follicle-stimulating hormone at a dose of 150–225 IU/day, according to age and BMI, starting at day 7 free of oral contraceptives. Daily injections of GnRh antagonist started when a dominant follicle measuring >14 mm was observed at ultrasound; in the case of corifollitropin alfa use, the antagonist was introduced on the fifth day in fixed protocol. Ovulation was triggered with a bolus of GnRH agonist (triptoreline, 0.2 mg)—the combination of a GnRh antagonist protocol with agonist triggering virtually eliminates the likelihood of ovarian hyperstimulation syndrome.⁴

Anti-Müllerian hormone assay

Serum AMH was determined by using the Immunotech AMH enzyme immunoassay (Beckman-Coulter Company).

Outcome measures

According to the regulations of the PGB, each pack of oocytes sent to recipients of ART procedures must contain at least six mature oocytes (meiosis MII oocytes). This threshold seems to be a desirable minimum to consider the woman candidate for a future upcoming donation.^1,5 With this work and given the high economic cost supported by the PGB with the oocyte donation program, we intend to determine the value of AMH capable of predicting a retrieval of at least 12 oocytes during the oocyte pick-up (OPU) in our population (enough oocyte number to satisfy at least two requests for oocytes). The main outcome measures were to determine whether AMH values are related to the degree of ovarian response.

Statistical analysis

Baseline characteristics (AMH, BMI, and age) and results related to ovarian response (total stimulation dose of gonadotropins required and days of stimulation) were evaluated and characterized. In addition, Pearson correlation coefficients were calculated to evaluate whether ovarian response (number of oocytes retrieved) was associated with AMH and the age of the donors. Variables associated with the number of oocytes retrieved (statistical significance was defined as P < .05) were included in a multivariate linear regression model to identify the unstandardized coefficients and 95% confidence interval for factors independently related to the oocyte retrieval rate. Receiver operating characteristic curves were constructed for each of the parameters tested to assess the sensitivity and specificity of AMH in predicting normal response (12 oocytes) to stimulation in oocyte donors.^1,5 All analyses were performed using STATA 14.2 (Statistics/Data Analysis, Copyright 1985-2015 StataCorp LP).

Results

A total of 48 candidates were included in this study (oocyte donors who underwent OPU). Of these 48 candidates, 34 completed the screening with ovarian reserve evaluation with AMH. Two oocyte donors were excluded because of OPU without oocytes. The oocytes from 29 donors were cryopreserved (ART cycles with cryopreserved oocytes) and from three donors were used in fresh ART cycles and cycles with cryopreserved oocytes. The characteristics of the sample analyzed and the ovarian response to the treatment are presented in Table 1. The mean age of the participants was 27 years. The ovarian reserve evaluation showed a mean AMH of 5.20 ng/mL. An average number of 16 oocytes was retrieved (12 mature oocytes MII). AMH levels showed a statistically significant positive correlation with the number of total oocytes retrieved (Fig. 1).

Table 1.

Baseline characteristics of oocyte donors and ovarian response.

N = 32 oocyte donors	Mean ± SD	Range	AMH correlation (r)
Age (y)	26.63 ± 3.97	20–33	−0.1756 (P = 0.3364)
BMI (kg/m2)	23.36 ± 2.91	18.69–29.62	−0.0422 (P = 0.8188)
AMH (ng/mL)	5.20 ± 4.67	1.31–25.81	—
Days of stimulation (days)	10 ± 2	6–16	−0.1427 (P = 0.4359)
Total dose of rFSH (IU)	1798 ± 529	975–2700	−0.2691 (P = 0.1363)
Oocytes (n)	16 ± 10	3–44	0.3828 (P = 0.0306)*
Oocytes MII (n)	12 ± 7	2–32	0.3233 (P = 0.0710)

BMI—body mass index, AMH—anti-Müllerian hormone, rFSH—recombinant follicle-stimulating hormone, IU—international units, oocytes MII—mature oocytes in the MII meiosis stage, SD—standard deviation, r—adjusted Pearson correlation coefficients between baseline AMH levels, age, BMI, days of stimulation, total dose of gonadotropins, and number of total and MII oocytes retrieved.

*P < 0.05.

Figure 1.

Adjusted Pearson correlation coefficient between baseline AMH levels and number of total oocytes retrieved (P < 0.05).

A threshold value of AMH = 3.2 ng/mL predictive of the retrieval <12 oocytes (areas under the curve, 0.7364; 95% confidence interval: 0.529–0.944) was identified by receiver operating characteristic curve. Using this cutoff, the normal response (12 oocytes) was predicted with a sensitivity of 77% and a specificity of 60% (Fig. 2).

Figure 2.

Receiver operating characteristic (ROC) curve for prediction of a retrieval of at least 12 oocytes during the oocyte pick-up in our population (enough oocyte number to satisfy at least two requests for oocytes).

Discussion

In the PGB oocyte donor population considered, the value of AMH shows a tendency (not significant) to decrease with age. This correlation has already been duly confirmed by other authors.^1,5 The small sample size is a limiting factor in the interpretation of these results and the methodology of laboratory determination of the value of AMH. Our laboratory uses the Beckman-Coulter Company methodology, but some studies have begun to show that this methodology may not be reliable and reproducible.¹⁰ Hence, the great heterogeneity of the values of the AMH was already seen in other series.⁵

AMH levels correlated significantly with the number of oocytes retrieved (r = 0.3828) and the AMH cutoff to predict a retrieval of at least 12 oocytes (3.2 ng/mL) showed a good capacity to discriminate an appropriate response in an oocyte donor candidate in our population (areas under the curve > 0.7). The AMH value also correlated positively with the number of collected MII oocytes, although the association was not significant. In the study by Martínez et al (2013), this association was significant, which may be due to the small size of our sample or differences in the methodology of selection of participants and in the objectives of that study (to determine the threshold of AMH capable of predicting a low response to stimulation for identify potential poor responders—women with <6 oocytes MII in the OPU).⁵

At our center, we routinely conducted AFC. However, the measurement is recorded at intervals not allowing us to retrospectively measure the exact number of antral follicles. It has been proven that AFC is a good marker in predicting the ovarian response to cycles of controlled ovarian stimulation in ART cycles (oocyte donors and infertility sufferers),^3,5 so in the future, we will apply in our protocols the registration of the nominal count of antral follicles.

The determination of AMH according to a reliable laboratory methodology seems to be a reproducible method of evaluating the ovarian reserve, with the advantage of not being influenced by the day of the menstrual cycle in which it is measured and not being a dependent operator,^7,8 which makes it a marker particularly useful in evaluating a population of oocyte donors.

In the case of the PGB, the measurement of AMH associated with the AFC may be a determining factor in the choice of the candidates to maximize the response to requests from beneficiaries who require donor oocytes to perform ART cycles.

Financial support and sponsorship

None.

Conflicts of interest

None.

This work was developed within the scope of The Portugal Clinical Scholars Research Training (PTCSRT) program (2017–2019). The PTCSRT program is a collaboration between Harvard Medical School and Fundação para a Ciência e a Tecnologia (FCT).