https://orcid.org/0000-0003-1762-6184
Abstract
Are meiotic and developmental competence of human oocytes from small (2–8 mm) antral follicles improved by applying an optimized IVM method involving a prematuration step in presence of C-Type Natriuretic Peptide (CNP) followed by a maturation step in presence of FSH and Amphiregulin (AREG)?
A strategy involving prematuration culture (PMC) in the presence of CNP followed by IVM using FSH + AREG increases oocyte maturation potential leading to a higher availability of Day 3 embryos and good-quality blastocysts for single embryo transfer.
IVM is a minimal-stimulation ART with reduced hormone-related side effects and risks for the patients, but the approach is not widely used because of an efficiency gap compared to conventional ART. In vitro systems that enhance synchronization of nuclear and cytoplasmic maturation before the meiotic trigger are crucial to optimize human IVM systems. However, previous PMC attempts have failed in sustaining cumulus-oocyte connections throughout the culture period, which prohibited a normal cumulus-oocyte communication and precluded an adequate response by the cumulus-oocyte complex (COC) to the meiotic trigger.
A first prospective study involved sibling oocytes from a group of 15 patients with polycystic ovary syndrome (PCOS) to evaluate effects of a new IVM culture method on oocyte nuclear maturation and their downstream developmental competence. A second prospective study in an additional series of 15 women with polycystic ovaries characterized and fine-tuned the culture conditions.
Fifteen women with PCOS (according to Rotterdam criteria) underwent IVM treatment after 3–5 days of highly purified human menopausal gonadotropin (HP-hMG) stimulation and no human chorionic gonadotropin (hCG) trigger before oocyte retrieval. A first study was designed with sibling oocytes to prospectively evaluate the impact of an IVM culture method: 24 h PMC with CNP + 30 h IVM with FSH and AREG, on embryo yield, in comparison to the standard (30 h) IVM clinical protocol (Group I, n = 15).
A second prospective study was performed in 15 women with polycystic ovaries, to characterize and optimize the PMC conditions (Group II, n = 15). The latter study involved the evaluation of oocyte meiotic arrest, the preservation of cumulus-oocyte transzonal projections (TZPs), the patterns of oocyte chromatin configuration and cumulus cells apoptosis following the 24 and 46 h PMC. Furthermore, oocyte developmental potential following PMC (24 and 46 h) + IVM was also evaluated. The first 20 good-quality blastocysts from PMC followed by IVM were analysed by next generation sequencing to evaluate their aneuploidy rate.
PMC in presence of CNP followed by IVM using FSH and AREG increased the meiotic maturation rate per COC to 70%, which is significantly higher than routine standard IVM (49%; P ≤ 0.001). Hence, with the new system the proportion of COCs yielding transferable Day 3 embryos and good-quality blastocysts increased compared to routine standard IVM (from 23 to 43%; P ≤ 0.001 and from 8 to 18%; P ≤ 0.01, respectively). CNP was able to prevent meiosis resumption for up to 46 h. After PMC, COCs had preserved cumulus-oocyte TZPs. The blastocysts obtained after PMC + IVM did not show increased aneuploidy rates as compared to blastocysts from conventional ART.
The novel IVM approach in PCOS patients was tested in oocytes derived from small antral follicles which have an intrinsically low developmental potential. Validation of the system would be required for COCs from different (larger) follicular sizes, which may involve further adjustment of PMC conditions. Furthermore, considering that this is a novel strategy in human IVM treatment, its global efficiency needs to be confirmed in large prospective randomized controlled trials. The further application in infertile patients without PCOS, e.g. cancer patients, remains to be evaluated.
The findings of this pilot study suggest that the efficiency gap between IVM and conventional IVF can be reduced by fine-tuning of the culture methods. This novel strategy opens new perspectives for safe and patient-friendly ART in patients with PCOS.
IVM research at the Vrije Universiteit Brussel has been supported by grants from: the Institute for the Promotion of Innovation by Science and Technology in Flanders (Agentschap voor Innovatie door Wetenschap en Technologie—IWT, project 110680); the Fund for Research Flanders (Fonds Wetenschappelijk Onderzoek–Vlaanderen—FWO, project G.0343.13), the Belgian Foundation Against Cancer (HOPE project, Dossier C69). The authors have no conflicts of interest.
Introduction
IVM is an ART that induces the meiotic maturation from prophase I to metaphase II, in vitro, of oocytes retrieved from small and mid-antral follicles of unstimulated or minimally stimulated ovaries (De Vos et al., 2016).
Inspired by research in animals, IVM was pioneered by Trounson et al. (1994) in patients with polycystic ovary syndrome (PCOS) and has been proposed as a more patient-friendly ART alternative to conventional IVF. Contrary to IVF, which can result in hyperstimulation syndrome (OHSS) and other risks in high responders, IVM is the only ART method with no cases of OHSS reported. Hence, patients with PCOS represent the major target population for IVM treatment (Child et al., 2002).
Despite substantial improvements in IVM practice in the past decade, IVM outcomes (maturation rate, embryo yield and implantation potential of IVM-derived embryos) still remain highly variable. This can in part be attributed to the heterogeneity of IVM protocols across fertility centers (Son and Tan, 2010). Indeed, ovulation triggering with hCG prior to oocyte retrieval is widely used as an important modification of the IVM protocol (as used in veterinary field) following a short course of gonadotropin treatment, to support follicle growth (Chian et al., 2000; Son and Tan, 2010). This approach is referred as ‘hCG triggered IVM or truncated IVF’ (De Vos et al., 2016) and has been shown to improve oocyte maturation rate in PCOS patients (Chian et al., 2000). However, robust evidence regarding the effect of hCG trigger on cumulative live birth rates is lacking. Furthermore, exposure of follicles <12 mm to an ovulatory trigger results in a heterogeneous collection of in vivo–matured metaphase (MII)-stage and MI-stage oocytes and germinal-vesicle (GV)-stage oocytes, which impedes standardizing IVM culture conditions and ICSI timing for the entire oocyte pool (Son and Tan, 2010).
A strategy using a short course of FSH (Walls et al., 2015) or highly purified (HP)-hMG stimulation and no hCG triggering (De Vos et al., 2011) yields a more homogenous cohort of oocytes—at the immature GV-stage—to undergo IVM culture. Using this approach, ~50% of the immature oocytes (retrieved from small and mid-antral follicles) mature in vitro (Guzman et al., 2012) and display features associated with acquisition of nuclear and cytoplasmic maturation (i.e. condensed chromatin configuration, transcriptionally silencing) (Sanchez et al., 2015).
Synchronization of meiotic and cytoplasmic maturation in antral oocytes arrested at the immature GV-stage remains a major challenge and is of fundamental importance for successful fertilization and pre- and post-implantation development (Coticchio et al., 2015).
High intra-oocyte levels of cyclic adenosine monophosphate (cAMP, produced by the oocyte and granulosa cells, is crucial to maintain the nearly fully-grown oocytes under meiotic arrest (Conti et al., 1998, 2002, Mehlmann, 2005; Sela-Abramovich et al., 2006). Research in animal models has indicated that a non-physiological drop of cAMP levels in the oocyte results in asynchronous nuclear and cytoplasmic maturation (Hyttel et al., 1997; Dieleman et al., 2002; Gilchrist and Thompson, 2007; Vaccari et al., 2008).
Previous IVM research has focused on the implementation of prematuration culture (PMC) systems (before inducing IVM) using cAMP phosphodiesterase (PDE3A) inhibitors (i.e. phosphodiesterase Type 3 inhibitors, PDE3-I) in order to prolong meiotic arrest and promote competence acquisition in human and animal oocytes (Nogueira et al., 2003a, 2006; Shu et al., 2008; Vanhoutte et al., 2009). Whereas this approach was efficient in preventing meiosis resumption, maintenance of cumulus-oocyte communication and transzonal projections (TZPs), crucial for oocyte competence acquisition (Gilchrist et al., 2008; Luciano et al., 2011; Lodde et al., 2013, Macaulay et al., 2016), appeared inefficient (Nogueira et al., 2003a, 2006; Shu et al., 2008). As a result, attempts to induce meiotic resumption by mimicking in vivo maturation have not been successful (Nogueira et al., 2003a, 2006; Shu et al., 2008).
Several studies in animal models (Zhang et al., 2010, 2011, 2015a, 2015b, 2017; Franciosi et al., 2014; Zhong et al., 2015) have demonstrated that C-type Natriuretic Peptide (CNP) maintains meiotic arrest of immature oocytes after removal from antral follicles. Moreover, CNP is currently considered a natural inhibitor for oocyte maturation (Zhang et al., 2010; Kawamura et al., 2011; Tsuji et al., 2012). Within the follicle, CNP, secreted by mural granulosa cells, binds the Natriuretic peptide receptor 2 (NPR2), expressed in the cumulus cells (Zhang et al., 2010; Tsuji et al., 2012), and induces the production of cGMP. Cyclic GMP enters the oocyte via gap-junctional communication (GJC) (Norris et al., 2009) and regulates the levels of cAMP by competing for the hydrolyzing activity of oocyte specific PDE3A (Degerman et al., 1997), thus maintaining oocytes under meiotic arrest.
Furthermore, the potential of CNP to improve oocyte competence and embryo quality in mammals has been demonstrated in large animal models (Zhang et al., 2015a, 2015b, 2017).
Inspired by Zhang et al. (2010) we recently demonstrated a remarkable improvement of IVM efficiency when CNP was added to a fine-tuned murine PMC/IVM system, developed for oocytes retrieved from unprimed prepubertal mice (Romero et al., 2016).
Here, we report the results of a prospective pilot study on sibling cumulus-oocyte complexes (COCs) aspirated from small antral follicles in 15 infertile patients with PCOS who underwent ART with a new IVM culture approach using PMC with CNP followed by IVM with FSH and Amphiregulin (AREG). In a complementary prospective study involving 15 women with polycystic ovaries we explored the limits of prematuration COC culture.
Materials and Methods
Study approval
This study was approved by the local ethics committee for human clinical studies of the University Hospital UZ Brussel of the Vrije Universiteit Brussel (project 2008/068), and by the Federal Commission for Medical and Scientific Research on embryos in vitro (Adv043/2012). Written informed consent was obtained from all PCOS subjects before inclusion in the study. The consent form included details pertaining to the donation of immature COCs to test a novel IVM method (PMC with CNP + IVM with FSH and AREG) and to assess efficacy and safety parameters in comparison to the current standard of practice.
Study design
To prospectively evaluate the impact of a new IVM culture system on oocyte maturation and embryo yield, 15 patients with PCOS (Rotterdam criteria, Rotterdam ESHRE/ASRM-sponsored consensus workshop group, 2004) undergoing IVM treatment were involved in a sibling oocyte study (Group I patients, n = 15).
Patients were recruited for this study if 30 or more antral follicles were visible on the last pelvic ultrasound scan before oocyte retrieval. The aspirates of an unselected subset of follicles were randomly allocated to the experimental arm of the study (PMC with CNP + IVM with FSH and AREG), whereas the majority of the COCs were allocated to the standard IVM protocol as part of patient's clinical treatment.
A complementary study to fine-tune the PMC conditions and optimize the PMC + IVM system was performed in a further series of 15 women with polycystic ovaries on ultrasound scan, i.e. oocyte donors with at least 12 antral follicles per ovary visible on a baseline ultrasound scan (Group II). This complementary study involved the assessment of oocyte meiotic arrest, maintenance of cumulus-oocyte TZPs, oocyte chromatin configuration, cumulus cell apoptosis and embryo development at two different PMC time points.
The number of participants (Groups I and II) and COCs in both pilot studies are shown in Fig. 1. Baseline patient characteristics for Groups I and II are presented in Table I.
Representative scheme showing the number of patients, COCs and embryos involved in the different analysis. In the sibling study, COCs from 15 patients were allocated to both systems, the new IVM method PMC with CNP followed by IVM with FSH and amphiregulin (AREG) and the standard IVM treatment, for the production of blastocysts (Group I). In a second study, 15 subjects (Group II) donated their COC's for the fine-tuning of PMC conditions, where 24 and 46 hour culture time points were considered. In this preliminary study, basic oocyte and cumulus cell features were evaluated by using staining protocols (as described in the Materials and Methods section). This study also included a preliminary evaluation on embryo development following PMC (at the two different time points indicated). CNP, C-Type Natriuretic Peptide; PMC, prematuration culture; COC, cumulus-oocyte complex. Some of the participants in Group II had insufficient oocytes and so their COCs were cultured only in one of the PMC time points. TZP, transzonal projections, NGS, next generation sequencing. N = number of participants (patients/subjects), n = number of COCs, oocytes or embryos analysed.
Baseline patient characteristics and cycle parameters.
| Group I | Group II | |
|---|---|---|
| No. of patients | 15 | 15 |
| BMI (kg/m2) | 25.9 ± 5.2 | 23.7 ± 4.8 |
| Age (y) | 28.9 ± 3.9 | 27 ± 3.6 |
| Gonadotropin priming: | ||
| Total injection dose (IU) | 680 ± 104 | 592 0.5 ± 144 |
| Days of stimulation | 3.4 ± 0.6 | 3.2 ± 0.4 |
| AFC (mean) | 53.7 | 37.8 |
| AMH (μg/l) | 12.9 ± 6.6 | 11.7 ± 6 |
| Baseline hormone values (Day 3) | ||
| E2 (ng/l) | 38.3 ± 10.9 | 43.2 ± 17.4 |
| Prog (μg/l) | 0.6 ± 0.3 | 0.6 ± 0.2 |
| FSH (IU/l) | 5.3 ± 1.6 | 6.1 ± 1.8 |
| LH (IU/l) | 6.6 ± 3.8 | 6.0 ± 3.0 |
| Hormones on day of OR | ||
| E2 (ng/l) | 686.9 ± 713 | 631.8 ± 753.3 |
| Prog (μg/l) | 0.8 ± 0.3 | 0.7 ± 0.3 |
| FSH (IU/l) | 7.4 ± 1.1 | 8.1 ± 1.8 |
| LH (IU/l) | 2.9 ± 2.5 | 3.5 ± 2.7 |
| Group I | Group II | |
|---|---|---|
| No. of patients | 15 | 15 |
| BMI (kg/m2) | 25.9 ± 5.2 | 23.7 ± 4.8 |
| Age (y) | 28.9 ± 3.9 | 27 ± 3.6 |
| Gonadotropin priming: | ||
| Total injection dose (IU) | 680 ± 104 | 592 0.5 ± 144 |
| Days of stimulation | 3.4 ± 0.6 | 3.2 ± 0.4 |
| AFC (mean) | 53.7 | 37.8 |
| AMH (μg/l) | 12.9 ± 6.6 | 11.7 ± 6 |
| Baseline hormone values (Day 3) | ||
| E2 (ng/l) | 38.3 ± 10.9 | 43.2 ± 17.4 |
| Prog (μg/l) | 0.6 ± 0.3 | 0.6 ± 0.2 |
| FSH (IU/l) | 5.3 ± 1.6 | 6.1 ± 1.8 |
| LH (IU/l) | 6.6 ± 3.8 | 6.0 ± 3.0 |
| Hormones on day of OR | ||
| E2 (ng/l) | 686.9 ± 713 | 631.8 ± 753.3 |
| Prog (μg/l) | 0.8 ± 0.3 | 0.7 ± 0.3 |
| FSH (IU/l) | 7.4 ± 1.1 | 8.1 ± 1.8 |
| LH (IU/l) | 2.9 ± 2.5 | 3.5 ± 2.7 |
Values are mean ± SD. AFC, Antral follicle count; AMH, antimüllerian hormone; OR, oocyte retrieval.
Baseline patient characteristics and cycle parameters.
| Group I | Group II | |
|---|---|---|
| No. of patients | 15 | 15 |
| BMI (kg/m2) | 25.9 ± 5.2 | 23.7 ± 4.8 |
| Age (y) | 28.9 ± 3.9 | 27 ± 3.6 |
| Gonadotropin priming: | ||
| Total injection dose (IU) | 680 ± 104 | 592 0.5 ± 144 |
| Days of stimulation | 3.4 ± 0.6 | 3.2 ± 0.4 |
| AFC (mean) | 53.7 | 37.8 |
| AMH (μg/l) | 12.9 ± 6.6 | 11.7 ± 6 |
| Baseline hormone values (Day 3) | ||
| E2 (ng/l) | 38.3 ± 10.9 | 43.2 ± 17.4 |
| Prog (μg/l) | 0.6 ± 0.3 | 0.6 ± 0.2 |
| FSH (IU/l) | 5.3 ± 1.6 | 6.1 ± 1.8 |
| LH (IU/l) | 6.6 ± 3.8 | 6.0 ± 3.0 |
| Hormones on day of OR | ||
| E2 (ng/l) | 686.9 ± 713 | 631.8 ± 753.3 |
| Prog (μg/l) | 0.8 ± 0.3 | 0.7 ± 0.3 |
| FSH (IU/l) | 7.4 ± 1.1 | 8.1 ± 1.8 |
| LH (IU/l) | 2.9 ± 2.5 | 3.5 ± 2.7 |
| Group I | Group II | |
|---|---|---|
| No. of patients | 15 | 15 |
| BMI (kg/m2) | 25.9 ± 5.2 | 23.7 ± 4.8 |
| Age (y) | 28.9 ± 3.9 | 27 ± 3.6 |
| Gonadotropin priming: | ||
| Total injection dose (IU) | 680 ± 104 | 592 0.5 ± 144 |
| Days of stimulation | 3.4 ± 0.6 | 3.2 ± 0.4 |
| AFC (mean) | 53.7 | 37.8 |
| AMH (μg/l) | 12.9 ± 6.6 | 11.7 ± 6 |
| Baseline hormone values (Day 3) | ||
| E2 (ng/l) | 38.3 ± 10.9 | 43.2 ± 17.4 |
| Prog (μg/l) | 0.6 ± 0.3 | 0.6 ± 0.2 |
| FSH (IU/l) | 5.3 ± 1.6 | 6.1 ± 1.8 |
| LH (IU/l) | 6.6 ± 3.8 | 6.0 ± 3.0 |
| Hormones on day of OR | ||
| E2 (ng/l) | 686.9 ± 713 | 631.8 ± 753.3 |
| Prog (μg/l) | 0.8 ± 0.3 | 0.7 ± 0.3 |
| FSH (IU/l) | 7.4 ± 1.1 | 8.1 ± 1.8 |
| LH (IU/l) | 2.9 ± 2.5 | 3.5 ± 2.7 |
Values are mean ± SD. AFC, Antral follicle count; AMH, antimüllerian hormone; OR, oocyte retrieval.
Stimulation protocol
For both study groups (I and II), administration of HP-hMG (Menopur, Ferring Pharmaceuticals SA, Aalst, Belgium) was started on cycle Day 3 of the menstrual period or a withdrawal bleeding. Ovarian stimulation was performed during 3 consecutive days, with a daily dose of 225IU, 225IU and 150IU HP-hMG, respectively. A pelvic ultrasound scan was performed in the morning of the third stimulation day to schedule the oocyte collection. If all follicles had a diameter of <6 mm, then 1 or 2 further days of HP-hMG stimulation days were added, but caution was taken for the diameter of the leading follicle not to exceed 10 mm.
Oocyte retrieval and culture procedure
COCs were retrieved 42 h after the last HP-hMG injection. In the standard IVM protocol, follicular aspirates were collected in Human Tubal Fluid (HTF) (IVF Basics® HTF HEPES, Gynotec B.V. Malden, the Netherlands) supplemented with heparin (5000 IU/ml, Heparin Leo, Leo Pharma, Belgium; final heparin concentration 20 IU/ml) and filtered through a cell strainer (Falcon®, 70 μm mesh size, BD Biosciences, CA, USA). After collection, COCs were washed and transferred to a four-well dish (Nunc; Thermo Fisher Scientific; MA, USA) containing IVM medium (IVM System, Medicult, Origio) supplemented with 75 mIU/ml HP-hMG (Menopur, Ferring, Saint-Prex, Switzerland), 100 mIU/ml hCG (Pregnyl, Organon, MSD, Haarlem, the Netherlands) and 10 mg/ml human serum albumin (HSA) (Vitrolife, Göteborg, Sweden). Cumulus-oocytes-complexes were cultured for 30 h in groups of 10 COCs per well in 500 μl IVM medium with oil overlay (Ovoil, Vitrolife) at 37°C under 6% CO2 in air.
For the new approach (PMC + IVM), COCs were collected in HTF supplemented with 50 μM 3-isobutyl-1-methylxanthine (IBMX) (Sigma, Schnelldorf, Germany) and heparin at 20 IU/ml. After collection, COCs were washed and transferred to a four-well dish, containing PMC medium (IVM System, Medicult, Origio) supplemented with 1mIU/ml rFSH (Puregon MSD, Australia), 5ng/ml Insulin, 10 nM estradiol (E2) (both from Sigma; Schnelldorf, Germany), 10 mg/ml HSA (Vitrolife, Göteborg, Sweden) and 25 nM CNP (Tocris Bioscience; Abingdon, UK). COCs were cultured in 500 μl of PMC medium, in groups of 10 COCs per well under oil for 24 h at 37°C, 6% CO2 in air.
Following 24 h of incubation in PMC media, COCs were thoroughly washed and transferred into Medicult IVM medium containing 5ng/ml Insulin, 10 nM E2, 100 ng/ml human recombinant Amphiregulin (rhAREG; Tocris Bioscience;) and 100 mIU/ml recombinant FSH and incubated for 30 h under same incubation conditions.
In both standard and experimental IVM protocols, naked or partially denuded oocytes (lacking >50% of cumulus cell layers) were not considered for culture.
For Group I, in total 264 COCs underwent standard IVM and 117 COCs were incubated in PMC media supplemented with CNP, followed by IVM with FSH and AREG.
ICSI and embryo culture
Thirty hours after IVM culture (following either standard IVM or PMC with CNP + IVM with FSH and AREG), oocytes were mechanically and enzymatically denuded from their cumulus layers under a stereomicroscope and oocyte maturation was assessed under the inverted microscope. Matured oocytes were inseminated using ICSI with partner sperm in the patient treatment arm (Joris et al., 1998) and donated sperm in the experimental arm. Fertilization was assessed 16–18 h post-insemination by the presence of two pronuclei. Fertilized oocytes and embryos were cultured in individual droplets of 25 μl medium with oil overlay (Ovoil, Vitrolife) until Day 5 (or Day 6) after ICSI.
Day 3 embryos were classified as transferable/ ‘good-quality embryos’ (GQE) according to the criteria described by Van Landuyt et al. (2013). Blastocysts were categorized according to Gardner et al. (1998) and Gardner and Schoolcraft (1999). Embryos generated in the experimental and treatment arms were evaluated by the clinical embryologists of UZ Brussel.
Embryo vitrification and embryo transfer—Standard IVM protocol
Day 3 embryos and blastocysts were vitrified and warmed according to the method described by Van Landuyt et al. (2011). The embryos obtained from the standard IVM protocol were frozen or transferred according to the policy of the Fertility Center at UZ Brussel: in patients with less than four good-quality embryos on Day 3 after ICSI, all good-quality embryos were vitrified at this stage for deferred transfer in artificial cycles (Ortega-Hrepich et al., 2013). In patients with ≥4 good-quality embryos, all embryos were cultured to Day 5 or 6.
Single fresh blastocyst transfer (SET) was performed in 5 patients; the remaining 10 patients had all their embryos vitrified on Day 3 after ICSI and underwent frozen embryo transfer in subsequent artificial cycles (Ortega-Hrepich et al., 2013).
Evaluation of blastocyst yield after PMC + IVM and standard IVM (Group I patients)
In the experimental arm, the final outcome was evaluated based on the amount of blastocysts eligible for vitrification.
Blastocysts after standard IVM were only available for those five (out of 15) cases with ≥4GQE D3. For the remaining 10 patients, blastocyst numbers were inferred by extrapolation (see supplementary information).
In total, 32 out of 64 GQED3 (50%) were further cultured up to blastocyst stage and 32 embryos were vitrified on D3. Based on extrapolation, it was estimated that 11 blastocysts could be obtained from the 32 frozen GQED3 of the clinical siblings’ data set, amounting to a final number of 22 blastocysts when including the actual number of frozen blastocysts.
Documentation of aneuploidy rate in blastocysts from the new IVM method
All blastocysts generated from PMC + IVM were vitrified for genetic and epigenetic analysis (safety endpoint). In total, 20 good-quality blastocysts were analysed using next generation sequencing (NGS) at The Babraham Institute (Cambridge, UK) using NextSeq500 High Output 100 bp Single End (Illumina, CA, USA).
Fine-tuning of ‘PMC’ conditions (Group II)
This group involved 15 oocyte donors with polycystic ovaries (at least 12 antral follicles on ultrasound scan). A similar approach as described earlier for oocyte retrieval, PMC with CNP and IVM with FSH and AREG was followed.
This preliminary study first focused on characterization of COCs/oocytes undergoing 24 and 46 h PMC in presence of CNP. Second, the impact of extended PMC (46 h compared to 24 h) followed by IVM (with FSH and AREG) on embryo yield was studied.
Morphology of human COCs
COCs from each IVM case were routinely imaged after retrieval, PMC and IVM, in order to record information about the patterns of cumulus proliferation and expansion. Live images were captured with a SC50 digital camera (Olympus, Japan) mounted on a Nikon Eclipse TE 300 inverted microscope (Nikon, Japan).
Assessment of cumulus-oocyte TZPs after PMC with CNP
Integrity of TZP following 24 or 46 h PMC in presence of CNP was assessed by fluorescent labelling of f-actin with Actin greenTM (Molecular Probes, Thermo Fisher Scientific) according to the manufacturer's instructions. Cumulus-oocyte-complexes were counter-stained with 1 μM ethidium homodimer-2 (Molecular Probes) for classification of chromatin configuration. Fluorescence signals were detected by confocal microscopy (IX70; Olympus, Japan) and images were acquired with Olympus Fluoview 2.1 software. Grayscale images of each COC were elaborated with FIJI software (available for download at http://pacific.mpi-cbg.de/wiki/index.php/Fiji).
Permanence and integrity of TZPs at the two different PMC time points were evaluated by morphological observation and compared between the two groups (N = 8 and N = 10 COCs for the 24 and 46 h groups, respectively).
Caspase-3/7 activation in cumulus cells after PMC
Apoptotic activity in cumulus cell layers of COCs exposed to CNP during 24 or 46 h PMC period was determined by short incubation with CellEvent™ Caspase-3/7 Green Detection Reagent (Molecular Probes) according to the manufacturer's instructions (9 COCs at each PMC time point). Cumulus-oocyte-complexes were fixed in 2% paraformaldehyde, counter-stained with 1 μM ethidium homodimer-2 and processed for confocal imaging as previously described. Three stacks were captured for each oocyte always at the same distance (middle section, 10 μm above and 10 μm below the center) and the number of caspase-3/7 positive cells, as well as total amount of cumulus cells per each stack, were determined by manual counting in Fiji (cell counter.jar plugin).
Evaluation of chromatin configuration in GV oocytes after PMC with CNP
COCs following 24 or 46 h PMC in presence of CNP, were processed for nuclear staining with 1 μM ethidium homodimer-2. Chromatin configurations of GV oocytes were assessed by confocal microscopy and classified as ‘dispersed’, ‘intermediate’ or ‘condensed’ on the base of chromatin condensation status around the nucleolus as reported by Sanchez et al. (2015). Frequency of distribution in the three categories was determined at the two PMC time points (18 COCs at 24 h and 31 COCs at 46 h) and compared to chromatin configuration after OPU (original data at retrieval from Sanchez et al., 2015).
Following 24 or 46 h PMC + IVM, fertilization of matured oocytes, embryo culture and embryo vitrification were performed as described earlier. The impact of PMC + IVM on embryo yield was assessed in 43 COCs in the 24 h and 55 COCs in the 46 h group, respectively.
Statistics
The influence of the PMC + IVM method on oocyte maturation and embryology outcomes, compared to sibling oocytes matured by standard IVM was assessed using a generalized linear mixed model for binomial responses, with patient as random factor and stimulation protocol as fixed factor by means of a probit link function. The effect estimates and their variance-covariance matrix were used to test the null hypothesis of no difference between culture systems.
Differences in integrity of TZPs at the two different PMC time points (24 and 46 h) were analysed by chi-square test. A t-test analysis of the percentages of caspase-3/7 positive cells per total cumulus cells (average of the three stacks), after Arc Sin transformation, was performed to compare two pools of COCs at the two different PMC time points. Statistical analysis with Fisher exact test was performed to highlight an eventual dependence between PMC time and chromatin configuration. Values of P < 0.05 were considered statistically significant.
Results
Stimulation characteristics and hormonal profiles
Group I: sibling oocyte study
Two days before oocyte retrieval all patients’ ovaries had more than 30 follicles visible on ultrasound scan (most of them measured between 2 and 6 mm, maximal follicular diameter did not exceed 8 mm) Table I. During oocyte retrieval, a total number of 773 follicles were punctured, 84% (n = 649) of which had a diameter between 1 and 6 mm and 16% (n = 124) between 7 and 10 mm. The mean oocyte recovery rate was 50%. A mean number of 18.9 COCs were randomly allocated to the patient's ART treatment (standard IVM) and 7.8 COCs were allocated to the PMC + IVM system. Blood sampling for hormone testing on the day of oocyte retrieval showed a high interpatient variability (Table I).
Group II: optimization of PMC conditions
The antral follicle count 2 days before oocyte retrieval showed the presence of between 19 and 74 follicles (per subject). The hormonal profile on the day of oocyte retrieval showed high interpatient variability (Table I).
Effect of the PMC + IVM system on oocyte competence and blastocyst yield (Group I: sibling oocyte study)
Figure 2 shows maturation outcomes of Group I oocytes (n = 117) cultured in the IVM system: 24 h PMC with CNP + 30 h IVM with FSH and AREG, versus oocytes (n = 264) undergoing 30 h standard IVM. Oocytes cultured in the PMC + IVM system showed a higher (P = 0.0001) maturation potential in comparison to standard IVM: 70% versus 48% polar body extrusion rate. Although fertilization rates did not differ between oocytes undergoing PMC + IVM or standard IVM (Fig. 2, rates over COC; Supplementary Fig. S1), the yield of Good-Quality Embryos on Day 3 per COC was nearly twice as high (Fig. 2 and Supplementary Fig. S1) in the PMC + IVM compared to standard IVM group: 43% versus 23% GQED3/COC, P = 0.0006. As result of the increased oocyte maturation in the PMC + IVM group patients had more top quality embryos available on Day 3 (ED3 Q1-top quality) and Good-Quality Blastocysts (GQ Blast) as compared to the standard IVM group (Fig. 2 and Supplementary Fig. S1): 23% versus 14% ED3 Q1-top quality/COC, P = 0.0461 and 18% versus 8%, GQ Blast/COC, P = 0.0116, Fig. 2.
Summary of the culture efficiency of the ‘standard IVM’ and the PMC + IVM method. Bar graph displays the main outcomes (values normalized to 100 oocytes) obtained after 24 h PMC with CNP + 30 h IVM with FSH and amphiregulin (AREG) applied to Group I oocytes (patients n = 15, COCs n = 117) versus 30 h standard IVM (sibling study, patients n = 15, oocytes n = 264). Numbers within the graph bars indicate actual numbers of COCs, metaphase II (mature) oocytes (MII), two pronuclei (fertilized oocytes, 2PN) good-quality embryo Day 3 (GQED3) and total top quality blastocysts, respectively. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001 (Statistical analysis performed as described in Materials and Methods section).
For the standard IVM group blastocyst production was determined by extrapolation (see Supplementary Information).
Blastocyst aneuploidy rate following the PMC + IVM system
The first 20 blastocysts available from the PMC + IVM strategy obtained from 10 patients, were analysed by NGS. In one blastocyst a partial trisomy of chromosome 5 was identified and two other blastocysts showed monosomy of chromosomed 4 and 18, respectively. The analysis revealed 12 XY and 8 XX embryos.
Fine-tuning of ‘PMC’ conditions (Group II)
In order to evaluate the ability of CNP to maintain meiotic arrest for an extended period of 46 h, a total of 63 COCs (from eight patients) retrieved from 2 to 9 mm follicles were incubated in PMC medium for 24 or 46 h. Overall, the assessment of meiotic resumption following PMC confirmed meiotic arrest at the GV-stage in 100% of COCs in which oocytes were fully covered with cumulus cells, at both 24 and 46 h. In a minority of oocytes, i.e. those not completely surrounded by cumulus layers, meiotic arrest at 24 and 46 h was maintained less efficiently: on average 83% (n = 22 COC; four patients) and 75% (n = 27 COC; five patients), respectively.
Presence of cumulus-oocyte TZPs following extended PMC supplemented with CNP (24 and 46 h)
Morphology of the COCs following PMC supplemented with CNP is shown in Fig. 3. Analysis of the TZPs in COCs indicated a tight connection between oocytes and cumulus cells at the end of PMC. Overall, fixed human COCs had radially-oriented TZPs at the oocyte-cumulus cell interface with foci of f-actin at the outer side of the zona pellucida (Fig. 4). When the two PMC time points were compared, TZPs were continuous and well-organized in a radial pattern after 24 h PMC in 100% of the analysed samples (n = 8); at 46 h, TZP amplification was observed due to massive cumulus cell proliferation with a partial loss of the radial pattern, possibly because of the overlap of several cumulus cell layers. Continuity of TZPs was observed in 70% of the analysed samples at 46 h (n = 10 COC). Statistical analysis of the morphological observations by chi-square test did not reveal a significant difference between the two PMC time points.
Representative images of human COCs morphology after retrieval, post PMC with CNP and post PMC with CNP followed by IVM with FSH and amphiregulin (AREG). Phase contrast images show two pools of COCs from a single patient exposed to 24 h PMC with CNP + IVM (a–c) and 46 h PMC with CNP + IVM (d–f). Panels a and d represent COCs at retrieval, as they were distributed in the two PMC groups after oocyte pick-up (OPU). Panels b and e, show compact COCs after 24 h (b) and 46 h (e) PMC with CNP. In panels c and f, is possible to appreciate cumulus expansion following 30 h IVM (with FSH and AREG) after 24 h (c) and 46 h (f) PMC culture with CNP, respectively. Original magnification, ×4. Scale bar: 100μm.
Cumulus-oocyte TZP are preserved after extended PMC with CNP. Immunofluorescent labelling of human COCs indicates maintenance of TZPs (green) integrity after 24 h (a–d, n = 8) and 46 h PMC (e–h, n = 10). F-actin staining (green) shows vertically-oriented TZPs on the tangential plane at the oocyte-cumulus cell interface (b.f). Cumulus cell nuclei and oocyte GV (ethidium homodimer-2, red) are shown in the merge (c,g). Original magnification, ×40, zoom 1.5. Scale bar: 50μm. Magnified TZPs details are depicted in panels d and h. Original magnification ×40, zoom 2.5. Scale bar: 50 μm. GV, germinal vesicle .
Influence of extended PMC (24 and 46 h) on cumulus cell apoptosis
On average, cumulus cell layers of COCs cultured for 24 h in PMC supplemented with CNP showed a very low degree of caspase-3/7 activation (0.81%, n = 9, from three subjects; Fig. 5). When COCs were exposed to extended PMC (46 h), a higher (P < 0.0001) incidence of apoptosis was observed (3.39%, n = 9, from three subjects; Fig. 5).
Caspase-3/7 activation in cumulus cell layers of human COCs following PMC with CNP.A. Detection of caspase-3/7 activation (green) in cumulus cells of human COCs after PMC with CNP by fluorescence imaging. The assay was performed on live cells at the end of 24 h (a–d) and 46 h (e–h) PMC with CNP. First panels at the left side show transmitted light images of representative COCs post 24 h (a) and 46 h (e) PMC with CNP; (b and f) caspase-3/7 activation in fluorescent bright green cumulus cells; Cumulus cell nuclei and oocyte GV (ethidium homodimer-2, red) are shown (c and g); merge (d.h). Original magnification, ×40, zoom 1.5. Scale bar: 50 μm. B. Statistical analysis of the percentage of caspase-3/7 positive cells per total number of cumulus cells per stack (average of the three stacks) and per condition by t-test (***P≤0.001, n = 9 in each group).
Detection of caspase-3/7 activation in cumulus cells layers of degenerating human COCs (positive control) showed a dramatic increase in apoptosis as 23.73% (n = 3, two subjects) of the cumulus cells were positively labelled (Supplementary Fig. S2).
Oocyte chromatin configuration patterns following extended PMC (24 or 46 h)
The frequency distribution over the three categories of chromatin configuration (dispersed, intermediate or surrounded) did not reveal a significant difference between the two PMC time points when compared to the status at retrieval (Fig. 6). However, the proportion of oocytes with both intermediate and condensed chromatin configurations tended to increase with an extending PMC: 72.2 and 87.1% in the 24 h (n = 18 COC, from four subjects) and 46 h group (n = 31 COC, from seven subjects), respectively.
Chromatin configuration of GV oocytes after PMC with CNP. (A) Representative images of magnified GV detail of human COCs after either 24 or 46 h PMC with CNP. Chromatin configurations of GV oocytes (ethidium homodimer-2; red) were assessed at retrieval (n = 79) and after 24 h (n = 18) or 46 h (n = 31) PMC with CNP by confocal microscopy and classified as ‘dispersed’ (a), ‘intermediate’ (b) or ‘condensed’ (c) on the base of chromatin condensation status and distribution around the nucleolus. Original magnification, ×40, zoom 1.5. Scale bar: 10μm. (B) Graph bar shows the frequency of distribution into the three chromatin configuration at the two PMC time points. Statistical analysis by Fisher exact test (ns). (*) Original data at retrieval from Sánchez et al., 2015.
Influence of PMC duration on blastocyst formation capacity
No significant differences were recorded in the embryology parameters in function of PMC culture time (Fig. 7 and Supplementary Fig. S3). The maturation and fertilization rates in oocytes cultured in PMC medium for 24 or 46 h were as follow: 74 and 76% MII rate and 58 and 55% 2PN rate of initially cultured oocytes, respectively. Embryo development was not affected by extension of the PMC time to 46 h (Fig. 7).
Impact of extended PMC with CNP (46 h) on maturation, fertilization and blastocyst yield. Bar graph displays the main outcomes (values normalized to 100 oocytes) of Group II oocytes after 24 h PMC with CNP (patients n = 5, oocytes n = 43) and 46 h PMC with CNP (patients n = 7, oocytes n = 55) followed by 30 h IVM. Maturation potential (metaphase II, MII), fertilization rate (2 pronuclei, 2PN), yield of good-quality embryos on Day 3 (GQED3) yield of top quality embryos and blastocyst did not differ between the two PMC time points. Numbers within the graph bars indicate actual numbers of COCs, MII, 2PN, GQED3 and blastocysts, respectively. Statistics by chi square test (ns).
Discussion
One of the major challenges when developing IVM systems is to tailor culture conditions to the stage-dependent oocyte needs. Although nuclear and cytoplasmic maturation are acquired in a step-wise fashion, these processes need to be coordinated to guarantee oocyte developmental competence.
It is generally accepted that meiotic arrest in fully-grown oocytes depends on the maintenance of high levels of cAMP (Conti et al., 1998, 2002). In the past, many systems have been proposed to improve oocyte IVM in both animal and human models including a PMC step preceding IVM, either using pharmacological compounds targeting phosphodiesterases to avoid degradation of cAMP (Thomas et al., 2002; Bilodeau-Goeseels, 2003; Nogueira et al., 2003a; Vanhoutte et al., 2008; Luciano et al., 2011) and/or compounds aiming to increase the cAMP load in cumulus cells and oocytes (Shu et al., 2008; Albuz et al., 2010; Zeng et al., 2013; Richani et al., 2014). Previous work in human COCs retrieved from small follicles using specific PDE3 inhibitors demonstrated that inhibition of PDEs can efficiently maintain meiotic arrest in vitro for up to 72 h (Nogueira et al., 2003a, b, 2006). However, reversing this artificial arrest by a positive hormonal stimulus (e.g. hCG or FSH in combination with EGF), even after diluting out the PDE3 inhibitor, was never successful. Microscopical observation of cultured COCs revealed that there was a progressive loss of connections between cumulus cells and oocyte during in vitro culture, hence preventing the transfer of signals among them (Nogueira et al., 2003a). In attempt to resolve this problem, more challenging culture systems using extracellular matrix support had to be designed; however, these approaches appear to be too complicated for routine applications (Vanhoutte et al., 2009).
Zhang et al. (2010) were the first to point to CNP as a principal regulator in the maintenance of oocyte meiotic arrest in antral follicles in mice. Since then, many reports from animal research indicated that CNP is able to sustain in vitro meiotic arrest of fully-grown oocytes for variable culture intervals: up to 24 h in mouse (Zhang et al., 2011) and for 6 or 8 h in COCs from mid-size antral follicles in goat and porcine (Zhang et al., 2015a,b) and bovine (Franciosi et al., 2014; Zhang et al., 2017). However, although some improvements in oocyte/embryo quality were reported when using CNP to temporarily inhibit spontaneous meiotic resumption in mouse oocytes (Zhang et al., 2010; 2011), most of these PMC systems targeted mid-to-large antral follicles containing nearly fully-grown oocytes after a treatment course with FSH. Using a more challenging approach, consisting of COC sampling from prepuberal mice (20 days old) unprimed with gonadotropins, our group applied PMC supplemented with CNP for up to 48 h, which was followed by IVM in presence of EGF-like factors. This system remarkably enhanced maturation and blastocyst development and the improvements were mainly attributed to the PMC step in presence of CNP (Romero et al., 2016).
The current study using human immature COCs in majority retrieved from follicles of 2–6 mm without pre-retrieval hCG trigger shows an equally significant benefit of the PMC + IVM system; the increased maturation capacity leads to a higher number of Day 3 embryos and good-quality blastocysts. These preliminary results in human also point to an important role of CNP in extending in vitro culture in order to support functions involved in competence acquisition (as revealed in the complementary PMC experiments).
COCs from 2 to 6 mm follicles are generally not utilized in human ART and their potential to develop into normal blastocysts following IVM has never been reported (Trounson et al., 2001).
Twenty of the first vitrified-warmed blastocysts obtained after PMC with CNP followed by IVM with FSH and AREG were assessed for chromosomal normality by an independent genetics laboratory using NGS. The aneuploidy rate as measured in complete blastocysts demonstrated no increased aneuploidy as compared to biopsied cells from blastocysts in age-matched conventional stimulation treatments used in ART (Franasiak et al., 2014). In view of potential epigenetic modifications that could be induced by prolonged in vitro culture, further analysis is currently ongoing using whole genome DNA methylation profiling of the PMC + IVM blastocysts (by post bisulfite adaptor tagging, PBAT analysis) in comparison to IVF control embryos and the best available reference data set of human blastocyst genome-wide DNA methylation (Okae et al., 2014).
The prospective sibling oocyte study done in 15 PCOS patients (Group I) was primarily chosen to explore the potential of a CNP/AREG IVM method to increase the competence of oocytes from small antral human follicles. Additional work using donated COCs from healthy donors with PCOS (Group II) aimed to further explore the potential of the PMC step. We hypothesized that extending the PMC from 24 to 46 h might even allow the successful maturation of oocytes originating from the smallest follicles. Similar to the mouse model (Romero et al., 2016), cumulus-oocyte TZPs were present after PMC with CNP. TZPs facilitate the formation of intercellular gap junctions (Albertini et al., 2001; Thomas and Vanderhyden, 2006) and play a role in the regulation of oocyte growth and meiotic competence acquisition (Carabatsos et al., 1998; Albertini et al., 2001). Work from others using alternative systems for PMC in human has revealed that TZPs (Nogueira et al., 2003a) and GJC (Shu et al., 2008) can be maintained for 6 h using a combined treatment of pharmacological compounds such as cilostamide and forskolin (Shu et al., 2008). The maintenance of TZPs in human COCs for up to 46 h in PMC supplemented with CNP (and in absence of pharmacological compounds) in the current study might contribute to the acquisition of oocyte competence. Nevertheless, further studies are required to confirm a functional status of GJC in COCs during PMC. Despite the fact that 46 h culture condition provided equally high oocyte maturation and blastocyst yields, a small, but significant increase of cumulus cell apoptosis was observed. Given that healthy follicles always contain a few apoptotic follicular cells it is likely that surpassing a critical threshold of apoptotic cells could preclude oocyte viability (Zeuner et al., 2003).
An important aspect associated with the maintenance of an intact cumulus-oocyte connection is the pattern of oocyte chromatin configuration. COCs retrieved from a similar pool of small follicles (2–8 mm) revealed that, at the moment of retrieval, around 50% of oocytes displayed condensed chromatin configuration and transcriptional silencing, indicating that these processes had been already initiated (Sanchez et al., 2015). Although it is still not clear how granulosa cells regulate chromatin remodelling and transcription in GV-arrested oocytes, several researchers have reported that these changes are crucial to confer oocytes with meiotic and developmental competencies (Zuccotti et al., 2002; De La Fuente, 2006; Lodde et al., 2007; Luciano et al., 2012, 2014). Studies in bovine indicate that modulation of these events through GJC partly involves cAMP-dependent mechanisms and that premature disruption of GJC during final oocyte development causes abrupt chromatin condensation and transcriptional repression (Luciano et al., 2011). Notably, studies in bovine oocytes have also revealed that CNP treatment in vitro delays meiotic resumption and sustains GJC (Franciosi et al., 2014).
In the current study, the proportion of oocytes with condensed chromatin was similar at retrieval and at the two PMC time points. Interestingly, a gradual transition from the dispersed to intermediate chromatin configuration was noticed in relation to PMC extension (up to 46 h culture). The current observations indicate that PMC media supplemented with CNP are able to keep human oocytes under meiotic arrest, maintain the integrity of cumulus-oocyte TZPs and sustain the dynamic changes in oocyte chromatin remodelling.
Additional factors such as E2 and oocyte secreted factors are also essential to increase the effectiveness of the CNP/NPR2 system (Zhang et al., 2010, 2011). Furthermore, the use of low FSH doses during culture of oocytes from small antral follicles promotes the efficiency of intercellular coupling between oocyte and cumulus cells (Luciano et al., 2011; Romero et al., 2016), while high FSH doses would induce a premature retraction of TZPs (Combelles et al., 2004). A molecular basis for the use of FSH to improve oocyte developmental competence has been provided by a study showing an interaction of FSH and the EGF-network to activate the PI3K/AKT cascade to regulate the translational program within the oocyte (Franciosi et al., 2016). A critical FSH dose range may exist to promote oocyte-cumulus connections and allow for the activation of PI3K/AKT, while avoiding initiation of the ovulatory cascade in cumulus cells.
Based on the previous findings by others that addition of EGF-like factors such as AREG to the maturation medium promote oocyte maturation and has been associated with a higher oocyte developmental competence (Zamah et al., 2010; Ben-Ami et al., 2011; Peluffo et al., 2012, Richani et al., 2013, 2014), this factor was selected together with FSH to induce maturation following PMC.
Overall, our findings support the combined role of factors used in this PMC + IVM system on the in vitro competence acquisition of human oocytes from small antral follicles which would otherwise have a very low meiotic and developmental potential.
It is plausible to compare the efficiency of this system as a mild alternative ART to conventional IVF in women with PCOS. Indeed, in view of the reduced gonadotrophin dose, minimal cycle monitoring and total avoidance of OHSS, IVM can foster a substantially reduced treatment burden (De Vos et al., 2011; Walls et al., 2015).
In the present pilot study, the clinical outcomes of the routine standard IVM (sibling) group demonstrate that Day 5 blastocyst transfer in the fresh cycle or a deferred transfer of vitrified/warmed Day 3 embryos can result in good live birth rates (see Supplementary Information). However, when compared to conventional IVF stimulation, blastocyst yield per treatment cycle is still lower after IVM (Walls et al., 2015). Indeed, the results of a study performed by Shapiro et al. (2011) demonstrated that, in a series of 91 cycles using the GnRH antagonist /FSH and GnRH Agonist trigger protocol, the blastocyst yield (blastocysts/COC) was 29.9%. As 18% of the COCs reached the blastocyst stage following PMC with CNP + IVM with FSH and AREG and 8% after routine sibling standard IVM, based on data of the current study, the efficiency gap between IVM and conventional IVF has been narrowed down substantially in the new PMC-IVM method.
Nevertheless, our study has some limitations: given that the crucial role of CNP involves generation of cGMP by cumulus cells it is a prerequisite that the oocyte aspiration procedure is non-disruptive for the cumulus-oocyte connections.
Considering PMC with CNP + IVM with FSH and AREG is a new strategy with future potential in human IVM treatment, the global efficiency needs confirmation in large prospective randomized controlled trials in comparison to currently used IVM systems (De Vos et al., 2016) and conventional IVF/ICSI.
Notwithstanding the remaining efficiency gap between PMC + IVM and conventional ART, the shorter treatment and the minimal therapeutic approach might already be appealing to PCOS patients.
In conclusion, the results of this pilot study involving CNP and AREG suggest that a significant further step has been made towards the development of a patient friendly and safe ART.
Supplementary data
Supplementary data are available at Human Reproduction online.
Acknowledgements
The authors thank the laboratory technicians, embryologists and paramedical staff at the Centre for Reproductive Medicine (CRG, UZ Brussel) for their kind collaboration and assistance. The authors acknowledge the advice of Wim Coucke from Scientific Institute of Public Health (Brussels) for performing part of the statistical analysis, the expertise of Gavin Kelsey from The Babraham Institute (Cambridge) for analysing IVM blastocysts by NGS, and Mrs Sandra De Schaepdryver for the editorial support.
Authors’ roles
F.S., S.R. and J.S. conceived and designed the study. S.R. and J.S. supervised the whole project. F.S., F.L., H.V.R. and S.R. performed culture procedures and experiments. F.S., F.L., H.V.R., S.R. and J.S. analysed and interpreted the data. M.D.V. recruited and managed the patients, and supervised the clinical activities. E.A. supervised the hormonal monitoring and safety analysis. G.V. supervised the routine IVM laboratory activities. All authors discussed the results and implications of the study. F.S., F.L. and J.S. wrote the manuscript. All authors revised, edited and approved the manuscript.
Funding
IVM research at the Vrije Universiteit Brussel has been supported by grants from: The Institute for the Promotion of Innovation by Science and Technology in Flanders (Agentschap voor Innovatie door Wetenschap en Technologie—IWT, project 110680); the Fund for Research Flanders (Fonds Wetenschappelijk Onderzoek–Vlaanderen—FWO, project G.0343.13), the Belgian Foundation Against Cancer (Human Ovary Preservation Expertise [HOPE] project, Dossier C69).
Conflict of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
References
Author notes
The authors consider that the first two authors should be regarded as joint First Authors.
