A fresh start for IVM: capacitating the oocyte for development using pre-IVM

Abstract

BACKGROUND

While oocyte IVM is practiced sporadically it has not achieved widespread clinical practice globally. However, recently there have been some seminal advances in our understanding of basic aspects of oocyte biology and ovulation from animal studies that have led to novel approaches to IVM. A significant recent advance in IVM technology is the use of biphasic IVM approaches. These involve the collection of immature oocytes from small antral follicles from minimally stimulated patients/animals (without hCG-priming) and an ∼24 h pre-culture of oocytes in an advanced culture system (‘pre-IVM’) prior to IVM, followed by routine IVF procedures. If safe and efficacious, this novel procedure may stand to make a significant impact on human ART practices.

OBJECTIVE AND RATIONALE

The objectives of this review are to examine the major scientific advances in ovarian biology with a unique focus on the development of pre-IVM methodologies, to provide an insight into biphasic IVM procedures, and to report on outcomes from animal and clinical human data, including safety data. The potential future impact of biphasic IVM on ART practice is discussed.

SEARCH METHODS

Peer review original and review articles were selected from PubMed and Web of Science searches for this narrative review. Searches were performed using the following keywords: oocyte IVM, pre-IVM, biphasic IVM, CAPA-IVM, hCG-triggered/primed IVM, natural cycle IVF/M, ex-vivo IVM, OTO-IVM, oocyte maturation, meiotic competence, oocyte developmental competence, oocyte capacitation, follicle size, cumulus cell (CC), granulosa cell, COC, gap-junction communication, trans-zonal process, cAMP and IVM, cGMP and IVM, CNP and IVM, EGF-like peptide and IVM, minimal stimulation ART, PCOS.

OUTCOMES

Minimizing gonadotrophin use means IVM oocytes will be collected from small antral (pre-dominant) follicles containing oocytes that are still developing. Standard IVM yields suboptimal clinical outcomes using such oocytes, whereas pre-IVM aims to continue the oocyte’s development ex vivo, prior to IVM. Pre-IVM achieves this by eliciting profound cellular changes in the oocyte’s CCs, which continue to meet the oocyte’s developmental needs during the pre-IVM phase. The literature contains 25 years of animal research on various pre-IVM and biphasic IVM procedures, which serves as a large knowledge base for new approaches to human IVM. A pre-IVM procedure based on c-type natriuretic peptide (named ‘capacitation-IVM’ (CAPA-IVM)) has undergone pre-clinical human safety and efficacy trials and its adoption into clinical practice resulted in healthy live birth rates not different from conventional IVF.

WIDER IMPLICATIONS

Over many decades, improvements in clinical IVM have been gradual and incremental but there has likely been a turning of the tide in the past few years, with landmark discoveries in animal oocyte biology finally making their way into clinical practice leading to improved outcomes for patients. Demonstration of favorable clinical results with CAPA-IVM, as the first clinically tested biphasic IVM system, has led to renewed interest in IVM as an alternative, low-intervention, low-cost, safe, patient-friendly ART approach, and especially for patients with PCOS. The same new approach is being used as part of fertility preservation in patients with cancer and holds promise for social oocyte freezing.

Graphical Abstract

Biphasic IVM is a simple variation of standard IVM consisting of a pre-IVM step, where the oocyte is deliberately arrested at the germinal vesicle stage with the aim of enhancing oocyte development, and an IVM step, together leading to improved embryo and pregnancy outcomes.

pre-maturation, biphasic IVM, capacitation IVM, oocyte developmental competence, minimal stimulation, IVM

Introduction

IVF was initially developed by Edwards and Steptoe by aspirating the mature oocyte from the dominant follicle in a natural cycle (Steptoe and Edwards, 1978). In an attempt to obtain more oocytes for fertilization, Sir Robert Edwards studied the immature oocytes retrieved from smaller antral follicles in different species and reported on their kinetics of IVM (Edwards, 1965). However, rather than pursuing IVM, from the 1980s onwards the ART community focused instead on optimization of ovarian stimulation and its monitoring as a means to generate multiple ova for IVF. The primary aim of today’s ART treatment strategy is to achieve a single healthy live birth, optimizing efficiency, and reducing time to birth. To reach this goal, ovarian stimulation typically needs to generate 12–18 oocytes for an optimal chance of a subsequent live birth (Law et al., 2021). The supernumerary embryos obtained by such a strategy are frozen for subsequent transfer cycles. In a country like Belgium, where up to six IVF cycles are reimbursed and where all starting treatments are centrally registered, a couple requires on average 2.5 stimulation cycles to take home a baby (De Neubourg et al., 2016). While complications with the current ovarian stimulation protocols have been notably reduced over the last decade, IVF treatment nevertheless still has a significant adverse impact on women’s daily lives. Data from the Belgian registry reveal that after three unsuccessful IVF attempts, the rate of abandonment of IVF is considerable, because of the burden of treatment (De Neubourg et al., 2021).

An alternative strategy, as originally presented by Edwards (1965), would be to collect oocytes from antral follicles in an unstimulated natural cycle, make these oocytes developmentally competent in vitro, and use them later, at the patient’s convenience, for embryo transfer (ET). Can IVM and its associated knowhow deliver such a promise? Today we have reached a level of technological sophistication in several relevant methodologies that could permit the implementation of such a low-intervention approach for infertile couples. The knowledge for this modern in vitro approach is founded on decades of exhaustive IVM research and clinical practice in a wide range of animal species: notably, advanced livestock breeding implemented these modern ART technologies many years ago, and ∼400 000 transfers are performed per year of cattle embryos generated from unstimulated IVM/IVF (Merton et al., 2003; Viana, 2020). However, even though human IVM is an older technology than conventional ovarian stimulation-IVF, clinical progress has been slow and has been marred by a confusing array of clinical protocols and definitions (De Vos et al., 2016, 2021), to the extent that it was only recently, in 2021, that it was declared non-experimental (Practice Committees of the American Society for Reproductive Medicine et al., 2021).

In the meantime, over the past ∼20 years, reproductive biology forged ahead with major advances from animal studies in the most fundamental aspects of ovarian and oocyte biology, including breakthroughs in our understanding of mechanisms of ovulation, regulation of meiosis, oocyte-somatic cell communication, oocyte metabolism, and oocyte developmental competence (key reviews: Downs, 2010; Conti et al., 2012; Coticchio et al., 2015; Clarke, 2018; Conti and Franciosi, 2018; Richani et al., 2021). A key step in the evolution of IVM has been the translation of this new basic knowledge into modern IVM protocols (Gilchrist, 2011). Collectively, this has led to ‘a fresh start for IVM’ with the development of pre-IVM methodologies which are incorporated into biphasic IVM. The objectives of this review are to examine the scientific rationale underpinning the development of pre-IVM strategies, with particular attention to the new approaches using c-type natriuretic peptide (CNP)-mediated pre-IVM, and implementation into clinical practice.

Current position of IVM in clinical practice

As a mild-approach ART that does not involve the development of large, pre-ovulatory follicles, and therefore carries essentially no risk of ovarian hyperstimulation syndrome (OHSS), IVM gained interest during the 1990s as an alternative ART for hyper-responders. Although several IVF clinics have developed an expertise in IVM over time, and in spite of acceptable live birth rates of 40% or more per IVM cycle in selected hyper-responders (Ho et al., 2019; Mackens et al., 2020), the uptake of IVM as an alternative, mild-approach ART is still very limited. Indeed, the original rationale for using IVM in the fertility clinic was to reduce the incidence of OHSS in those at increased risk, specifically women with PCOS. With the development of modern protocols for ovarian stimulation, including the use of a GnRH agonist trigger in GnRH antagonist protocols and embryo vitrification (Mourad et al., 2017), the incidence of severe OHSS has almost been eradicated. Only hyper-responders are still susceptible to hormonal side effects, including mild or moderate forms of OHSS, when they undergo ovarian stimulation for ART, despite use of OHSS risk mitigation strategies (Shrem et al., 2019). In rare cases, hyper-responders may develop ovarian torsion, a serious complication requiring urgent surgical intervention, as a result of ovarian stimulation (Vesztergom et al., 2021).

Other than being a safer alternative to conventional ART, IVM meets the recent trend towards less invasive and less expensive fertility treatments (Wessel et al., 2023) as there is no need to monitor ovarian response in IVM cycles that use significantly less gonadotrophins than traditional IVF. Hence, the cost of an IVM cycle is between 54% (Rose et al., 2014) and 66% (Braam et al., 2021) that of a traditional IVF cycle. However, as IVM is less efficient than IVF, the cost/baby to the patient depends on the extent to which ART costs are publicly subsidized in different countries. Braam et al. (2021) show that biphasic IVM is more cost-effective than IVF at a willingness-to-pay of up to €18 000 for an additional child, whereas above €18 000 IVF becomes more cost-effective. IVM may be a more acceptable ART for those patients who struggle to combine fertility treatment with the pressures of work and life. This struggle may contribute to distress related to IVF treatment and may lead to treatment termination before a successful pregnancy is achieved (Gameiro et al., 2012).

A further impediment to more widespread uptake of IVM is the limited evidence regarding its efficiency. So far only two high-quality randomized controlled trials (RCTs) have compared reproductive outcomes from one cycle of IVM with those from one cycle of conventional ART, showing lower cumulative ongoing pregnancy or live birth rates, mainly as a result of a reduced number of available embryos from IVM cycles (Vuong et al., 2020a; Zheng et al., 2022). In view of this, one may conclude that, with current IVM systems, proper selection of suitable patients should be the cornerstone of a clinical IVM program. Women with severe PCOS, as defined by the Rotterdam criteria, are the best candidates for IVM as they yield sufficiently high numbers of immature oocytes to make up for the inherently lower efficiency of IVM compared with standard IVF (Fadini et al., 2011; Guzman et al., 2013b; Seok et al., 2016). Such PCOS and high antral follicle count (AFC; i.e. >24 follicles in both ovaries; Broekmans et al., 2010) patients suitable for IVM make up ∼15% of the total ART population, as evidenced from the recent large RCT on biphasic IVM versus IVF where 5892 ART patients were screened and 917 patients were eligible for IVM (Vuong et al., 2020a). Other appropriate indications for IVM in routine clinical practice include: fertility preservation in women diagnosed with cancer who need systemic cancer treatment with a high risk of gonadotoxicity (Grynberg et al., 2016); and resistant ovary syndrome (ROS), a rare condition characterized by hypergonadotropic anovulation and absence of ovarian response to high doses of gonadotrophins as a result of a genetic or immune-mediated dysfunction of the FSH or LH receptor (LH-R) (Galvao et al., 2018; Le et al., 2021).

The limitations of traditional IVM approaches

Like conventional IVF, there are a number of different variations on IVM, both in clinical practice and laboratory protocols. Definitions and a brief explanation of terminologies are provided in Table 1. As the pros and cons of the various clinical protocols in relation to the range of patient indications are reviewed in detail elsewhere (De Vos et al., 2021; Gilchrist and Smitz, 2023), here we briefly examine the two main protocols in the context of the evolution of IVM practice.

Table 1.

IVM protocol definitions and terminologies.

Protocol/term	Definition/terminology	Key human study
Standard IVM	Maturation in vitro of immature GV-stage intact COCs in one step to MII, from unstimulated or FSH-primed patients.	Edwards (1965)
hCG-triggered IVM	Maturation of COCs in vivo and in vitro in one step to MII, from patients triggered with hCG or a GnRH agonist.	Chian et al. (1999)
Natural cycle IVF/M	A variation of hCG-triggered IVM for cycling women where the dominant follicle is tracked and triggered with hCG/GnRH agonist when >16mm	Chian et al. (2004)
Truncated IVF	Maturation of COCs in vivo and in vitro in one step to MII, from patients primed for several days with FSH and triggered with hCG or a GnRH agonist.	Fadini et al. (2009)
Ex vivo IVM or OTO-IVM	Maturation in vitro of immature GV-stage intact COCs collected from ovarian tissue ex vivo, typically without any prior gonadotrophin stimulation.	Prasath et al. (2014)
Pre-IVM	The first phase of biphasic IVM where intact COCs are deliberately arrested at the GV-stage	Nogueira et al. (2003a)
Biphasic IVM	Maturation in vitro of immature GV-stage intact COCs in two steps, from unstimulated or FSH-primed patients. Intact COCs are GV-arrested in step one and matured in step two.	Nogueira et al. (2006)
CAPA-IVM	A biphasic IVM system entailing a CNP-mediated pre-IVM step and an amphiregulin- and FSH-induced IVM phase.	Sanchez et al. (2017)
Rescue IVM	Maturation in vitro of cumulus denuded GV-stage oocytes collected from stimulated conventional IVF/ICSI cycles.	Veeck et al. (1983)

CAPA IVM, capacitation IVM; CNP, c-type natriuretic peptide; COCs, cumulus–oocyte complexes; EGF, epidermal growth factor; GV, germinal vesicle; MII, metaphase II; OTO IVM, ovarian tissue oocyte IVM.

Challenges with standard IVM

In standard IVM, the cumulus–oocyte complex (COC) is aspirated from follicles (2–10 mm), which usually have been supported by some days (generally 2–5 days) of FSH pre-treatment (150–250 IU daily), and no hCG trigger is used. The retrieval of COC from the follicles <10 mm is planned 42–46 h after the last FSH injection. These COCs are cultured for 24–36 h in IVM medium supplemented with gonadotrophins and often containing human homologous serum or serum substitute products or human serum albumin. Clinical pregnancy rates varied between 9.4% and 35% (Mikkelsen and Lindenberg, 2001; Fadini et al., 2009; De Vos et al., 2011; Guzman et al., 2012). The higher clinical pregnancy rates were obtained only by applying deferred ETs in artificial cycles; fresh ET in standard IVM delivers poor results owing to an insufficient endometrial priming (Ortega-Hrepich et al., 2013; Walls et al., 2015). A large prospective randomized study from China in unstimulated patients with PCOS, using serum-free IVM media and transferring a single vitrified/thawed blastocyst, reported an ongoing pregnancy rate of 22% after the first transfer, and 37% of cycles did not reach blastocyst transfer (Zheng et al., 2022). Therefore, even after deferred ET the efficiency of standard IVM in patients with PCOS without any gonadotrophin treatment is significantly inferior to conventional IVF after the first ET. Nevertheless, the standard IVM approach completely eliminated OHSS in patients with PCOS and was safe in a large retrospective analysis performed in a single center (Mackens et al., 2020).

Challenges with hCG-triggered IVM and natural cycle IVF/M

Triggering IVM patients with hCG was first reported in 1999 (Chian et al., 1999) and has been used widely since. A protocol variation thereof is natural cycle IVF/M used in cycling women who are triggered with hCG or GnRH agonist when the dominant follicle is >16 mm (Chian et al., 2004; Teramoto et al., 2016; Table 1). These protocols have the advantages of relatively high oocyte maturation rates of ∼70% (Zheng et al., 2012) but the disadvantage of generating oocytes with a mixed meiotic status at oocyte retrieval (Teramoto et al., 2016), necessitating at least two rounds of ICSI per retrieval, increasing burden on the laboratory. The original rationale for the protocol was that hCG-triggering would hasten meiotic resumption and thereby improve oocyte maturation rates and possibly aid oocyte retrieval (Chian et al., 2000), however, the logic of triggering small antral follicles with LH analogues has always been questioned. In human antral follicles (dissected out of ovaries for ovarian tissue preservation) the LH-R was more highly expressed in granulosa cells from smaller antral follicle stages than was previously believed. Jeppesen et al. (2012) found low amounts, but functional LH-R, in the majority of follicles of 5–6 mm diameter. Expression of FSH receptors and LH-R on granulosa cells increases in parallel with follicle diameter. In light of this finding, it can be better understood that when a massive hCG dose is given to a cohort of small pre-dominant follicles, partial, and/or inappropriate maturation effects were found in COCs (Girard et al., 2015). Hence, the maturing effects of hCG will vary in accordance with the diameters of follicles in the cohort: for the smallest follicles the hCG trigger will be insufficient to complete normal oocyte maturation. Rather a ‘spectrum’ of maturation effects will be observed depending on the level of LH sensitivity of each follicle. In the smaller follicles containing oocytes surrounded by granulosa not yet expressing LH-R (only LH-R in theca), with a still immature oocyte cytoplasm, hCG induces inappropriate signaling, compromising developmental competence, and leading to interruption of the essential communication between oocyte and cumulus (Girard et al., 2015).

With the benefit of 20 years hindsight from clinical practice using hCG-triggered IVM, there is no conclusive evidence that hCG-triggering has any effect on pregnancy, miscarriage, or live birth rates in IVM (Reavey et al., 2016). A large retrospective cohort study from China on 324 patients with PCOS treated either with standard IVM without any stimulation or triggered with hCG only, demonstrated no effect of the hCG trigger on oocyte maturation rate, number of embryos available, clinical pregnancy rate, and live birth rate (Lin et al., 2020). In a small RCT of patients with PCOS treated with or without hCG-triggering, the hCG trigger significantly improved oocyte nuclear maturation but had no effect on any other clinical outcomes including pregnancy and live birth rates (Zheng et al., 2012). Similarly, an RCT by Sonigo et al. (2020) investigating use of hCG-trigger IVM for urgent fertility preservation in normo-ovulatory patients with breast cancer did not demonstrate any difference in number of cryopreserved oocytes between no hCG, hCG-trigger, and GnRH-agonist trigger. From a comprehensive review of IVM clinical studies, there is no clear benefit of using an hCG-trigger in women with PCOS (n = 1812) on oocyte maturation or fertilization rates, and the clinical pregnancy rate was only 26% compared to 39.7% with FSH-priming but without hCG trigger (n = 315 cycles; Sauerbrun-Cutler et al., 2015). Given the lack of evidence of clinical benefit from hCG-triggering for IVM and the added burden it causes on the laboratory, there has been a recent movement of some centers away from hCG-triggering to trial new directions in IVM practice.

Folliculogenesis: atresia, gonadotrophins, and the developing oocyte

IVM is a technology that can use COCs collected from any size of antral ovarian follicle although, since in general terms oocytes progressively gain developmental competence with increasing follicle diameter, follicle size is an important determinant of IVM outcomes (Pavlok et al., 1992; Gilchrist et al., 1997; Hagemann et al., 1999). IVM practice in human and domestic animal ART programs capitalizes on the fact that mammalian ovaries at all cycle stages have numerous antral follicles at various stages of growth and atresia, containing viable germinal vesicle (GV)-stage COCs. In domestic animal species, 85% of antral follicles in an ovary at any time of the cycle show signs of atresia (Driancourt et al., 1991; Pavlok et al., 1992). It is common practice in livestock breeding to only perform IVM using COCs collected ex vivo from ovarian tissue from carefully selected small follicle diameters of 3.0 to 5.0 mm (Lequarre et al., 2005). COCs from such follicles are viable, do not contain apoptotic cells and lead to satisfactory blastocyst production. During the natural cycle in mono-ovulatory species, the dominant follicle develops while the subordinate follicles become progressively atretic (McNatty et al., 1979; Gougeon, 1986). The dynamics of ovarian follicle growth in humans has been meticulously mapped. Landmark studies with high-definition ultrasound from Baerwald and Pierson (Baerwald et al., 2003a) in the natural menstrual cycle revealed the existence of follicular wave patterns in women, as has been thoroughly documented in cow ovaries. Baerwald’s observations indicate that the small antral follicles, in the follicular phase (before the leading follicle reaches 9 mm) and in the luteal phase (before CL regression), contain COCs without major signs of atresia (Baerwald et al., 2003b; Baerwald and Pierson, 2020). Nonetheless, the pre-dominant follicular milieu has not yet been extensively studied in humans from an endocrine standpoint, nor its interaction with the immature COC. Recent efforts from fertility preservation centers using total ovarian tissues from cancer patients have started to provide detailed insights into follicular contents, including the study of receptors and proteins in the different cellular compartments of the follicle (Jeppesen et al., 2012; Guzman et al., 2013a; Sanchez et al., 2015; Kristensen et al., 2017, 2019).

FSH and LH are the principal survival factors for follicles that initiate formation of the antral cavity (when follicle diameter reaches 0.4 mm in humans). Baseline gonadotrophin levels are required to enable follicles to grow up to 2–5 mm diameter. It is the cohort of antral follicles of 2–5 mm diameter that requires the inter-cycle increase of FSH for its further growth (reviewed; Gougeon, 1996). Any follicles that reach 2–5 mm but miss the threshold FSH level will demonstrate atretic changes. The large majority of follicles >6 mm in a natural cycle are hence in various stages of atresia. In most human IVM studies so far, the follicle cohort targeted for collection has been between 6 and 10 mm, precisely where atretic changes are evident. Hence, the logical solution from practitioners was to prime the ovary with a few days of FSH preceding IVM retrieval to induce a healthier follicular environment for the COC (Suikkari et al., 2000; Fadini et al., 2009).

Rather than addressing a group of follicles of Class 5 and 6 (Gougeon classification) that already have a high rate of atresia, it makes sense to retrieve COCs from follicles <5–6 mm where apoptosis is much lower (Gougeon, 1986). Landmark studies from McNatty et al. (1979), who studied surgically removed whole human ovaries from women aged 25 to 49 years, reported that the average number of antral follicles of ≥4 mm diameter in both ovaries was 14 or less. The next landmark study on human ovarian tissue comes from Gougeon (1986), who described follicle classes 5 and 6 in particular, with a high proportion of atretic follicles (58–77%). Yuan and Giudice (1997) analyzed the frequency of apoptosis in follicles from normo-ovulatory women and demonstrated that it was maximal in the class of small antral follicles (2–5 mm). Mikkelsen et al. (2001) evaluated the occurrence of apoptosis in follicular aspirates from small antral follicles from unstimulated normal women and from gonadotrophin-stimulated normo-ovulatory and patients with polycystic ovaries (PCO), revealing a median rate of apoptosis of 46%. Ovarian stimulation for 3 days with recombinant FSH (150 IU/day), starting 3 days after onset of menses, reduced this rate of apoptosis by nearly 50% in normo-ovulatory women, but not in PCO patients with PCOS. These important studies form the basis of FSH-priming regimes (Mikkelsen and Lindenberg, 2001) commonly used in IVM clinical practice today.

Biological rationale for pre-IVM

New perspectives on mechanisms regulating oocyte maturation in vivo

Relevant to the development of pre-IVM approaches has been the discovery of the identity of the ‘oocyte maturation inhibitor’ characterized in the 1970s (Tsafriri et al., 1976), as CNP (Zhang et al., 2010). CNP, produced by the mural granulosa cells in particular, acts on the guanylyl cyclase natriuretic peptide receptor 2 (NPR2) present on the surface of both mural granulosa cells and cumulus cells (CCs), triggering the synthesis of cGMP (Zhang et al., 2010). cGMP diffuses through gap junctions to reach the oocyte and inhibits phosphodiesterase 3 (PDE3; Norris et al., 2009; Vaccari et al., 2009). cAMP is maintained at a high level in the oocyte by G-protein coupled receptor 3 activation (Mehlmann et al., 2002) and by supply of cAMP from CCs (Thomas et al., 2002; Webb et al., 2002), where cAMP inhibits maturation-promoting factor to maintain oocyte meiotic arrest at the GV stage (Fig. 1). The LH surge leads to a down-regulation of NPR2 and CNP and a rapid decrease in cGMP diffusion into the oocyte, leading to an activation of the oocyte’s phosphodiesterase type-3 (PDE3), a decline in cAMP, thereby triggering GV breakdown (GVBD) (Egbert et al., 2014; Shuhaibar et al., 2015; Zhang et al., 2010).

Figure 1.

Regulation of oocyte meiotic arrest in vivo. The ovarian follicle maintains mammalian oocyte meiotic arrest via the CNP-cGMP-cAMP network. CNP activation of its receptor NPR2 induces cGMP production, which is supplied to CCs and the oocyte to inhibit the oocyte’s PDE, maintaining sufficient intra-oocyte cAMP to prevent GVBD. The CC supply of cGMP alone is insufficient to maintain meiotic arrest. Solid lines indicate active pathways, dashed lines indicate downregulated processes during meiotic arrest. Illustration by Dr Emily Frost. CNP, c-type natriuretic peptide; CC, cumulus cell; GVBD, germinal vesicle breakdown; NPR2, natriuretic peptide receptor 2; GPR3, G-protein coupled receptor 3; MPF, maturation promoting factor; PDE3, phosphodiesterase type-3.

A second seminal discovery important to the development of modern IVM systems, is that the ovulatory LH surge does not act directly on CCs or the oocyte, which generally do not express LH-R, but rather that LH induces a cascade of epidermal growth factor (EGF)-like peptides, namely amphiregulin, epiregulin, and betacellulin, which lead to ovulation (Park et al., 2004). These EGF-peptides act on CCs causing withdrawal of the CCs trans-zonal processes (TZPs; Abbassi et al., 2021), loss of CC–oocyte gap-junctional communication (GJC) and cumulus expansion, facilitating meiotic resumption and leading to ovulation (reviewed; Conti et al., 2012; Richani and Gilchrist, 2018). COCs in small antral follicles have an under-developed EGF receptor signaling system (Ritter et al., 2015), which develops under the control of FSH (El-Hayek et al., 2014) in conjunction with oocyte-secreted growth differentiation factor 9 (GDF9) and bone morphogenetic protein 15 (BMP15; Su et al., 2010; Ritter et al., 2015; Sugimura et al., 2015). Development of EGF receptor signaling functionality on CCs is crucial for the normal physiological induction of meiotic resumption by the EGF-like peptides (Richani and Gilchrist, 2018; Abbassi et al., 2021).

Pre-IVM capacitates the oocyte for development

In the interests of furthering treatment opportunities for patients, it is incumbent on research and clinician scientists to use this new knowledge of mechanisms regulating oocyte maturation in vivo to improve IVM; to make it more physiological by attempting to mimic in vitro the newly discovered in vivo processes. A central theme that has emerged from this large research knowledge base in animals is that spontaneous oocyte maturation (i.e. the precocious resumption of meiosis upon COC removal from the follicle) should be avoided, necessitating some form of pre-IVM culture that actively prevents GVBD in vitro (Fig. 2) (Gilchrist, 2011). There are two basic rationales for pre-IVM: first, spontaneous IVM leads to an immediate loss of oocyte-CC communication, including withdrawal of CC TZPs from the oocyte and loss of CC–oocyte GJC (Luciano et al., 2004; Thomas et al., 2004a). Hence the oocyte’s elaborate somatic cell support network is lost prematurely, adversely affecting oocyte developmental competence. Second, IVM oocytes are typically collected from small to mid-sized antral follicles, which are not ready to be ovulated and support embryo development. These oocytes are still developing—they are still transcriptionally active and developing the molecular machinery needed to support future embryo development (Fair et al., 1997; Lodde et al., 2008). Pre-IVM seeks to continue the oocyte’s developmental programming in vitro that would otherwise be continuing in vivo—and nurturing of the oocyte by CCs is central to this process. Hence, pre-IVM culture is an oocyte ‘development’ phase (graphical abstract) intended to better capacitate the oocyte (Hyttel et al., 1997) in vitro for subsequent embryo development (see section ‘Effects of biphasic IVM on the COC’). A further advantage of pre-IVM is that it enables a controllable resumption of oocyte meiosis when the COC is moved into the IVM phase. As the oocyte is deliberately meiotically arrested up to this point, it allows the judicious use of meiotic-inducing agents, such as the EGF-like peptides, to control the timing of meiotic resumption as opposed to spontaneous meiotic resumption, which occurs without the use of a pre-IVM step. A pre-IVM step followed by IVM together make up a biphasic IVM culture system (Graphical abstract; Funahashi et al., 1997).

Figure 2.

Biphasic IVM culture system. Biphasic IVM protocols consist of two steps in vitro: a pre-IVM step (typically ∼24 h) intended to enhance GV oocyte development; and an IVM step (typically ∼30 h in human) where the oocyte meiotically matures. In the pre-IVM phase, the oocyte is deliberately arrested at the GV stage in vitro using meiotic inhibitors. The intact COC structure is preserved enabling continued CC support of oocyte development via gap junctions and provision of CC regulators in the medium. In the subsequent IVM phase, meiotic resumption is induced by factors such as EGF-like peptides and FSH. Thereafter MII oocytes are treated as per conventional IVF. CC, cumulus cell; COC, cumulus–oocyte complex; EGF, epidermal growth factor; GV, germinal vesicle; GVBD, germinal vesicle breakdown; MII, mature metaphase II oocyte; EGF, epidermal growth factor; PGE2, prostaglandin E2; CNP, c-type natriuretic peptide; GDF9, growth differentiation factor 9; BMP15, bone morphogenetic protein 15; PDEi, phosphodiesterase inhibitor; E2, oestradiol.

Pre-IVM methodologies

cAMP-mediated pre-IVM

The notion that spontaneous oocyte meiotic resumption in vitro could be prevented by modulating the cyclic nucleotides cAMP and cGMP was already well established in the 1970s (Cho et al., 1974; Magnusson and Hillensjo, 1977; Dekel and Beers, 1978). However, it was not until the late 1990s that the effect of deliberately inhibiting oocyte meiotic maturation with high levels of cAMP at the start of oocyte culture, before moving oocytes into an inhibitor-free IVM phase (biphasic IVM; Fig. 2), was first assessed in terms of embryo outcomes (Funahashi et al., 1997; Luciano et al., 1999). These were innovative ideas in IVM at the time. In the following two decades, there was a plethora of animal studies investigating cAMP-mediated IVM culture systems, particularly in domestic species, motivated by veterinary clinical and commercial applications (reviewed; Leal et al., 2018). The latter studies inspired also several important studies using human oocytes (Nogueira et al., 2006; Vanhoutte et al., 2007; Shu et al., 2008; Zeng et al., 2013; Spits et al., 2015). Many different approaches to using cAMP modulators in IVM have been investigated (reviewed; Gilchrist et al., 2016; Richani and Gilchrist, 2022). These include: in a biphasic IVM mode, using PDE3 inhibitors, such as cilostamide, milrinone or Org9935, in pre-IVM before wash-out, and an inhibitor-free IVM phase (Nogueira et al., 2003b; Luciano et al., 2011)—this protocol protects existing COC cAMP levels; in a monophasic IVM mode, also using PDE3 inhibitors but simultaneously in the presence of meiotic inducing ligands, such as FSH or EGF-like peptides, which eventually override the effect of the inhibitors to induce meiosis during IVM (Downs et al., 1986; Thomas et al., 2004b)—this system leads to moderate increases in COC cAMP; and in a biphasic IVM mode where high levels of cAMP are either supplied exogenously (e.g. dibutyryl cAMP; dbcAMP; Funahashi et al., 1997; Somfai et al., 2003) or are induced by pharmacological agents such as invasive adenylate cyclase (Luciano et al., 1999; Guixue et al., 2001) or forskolin (Ali and Sirard, 2005; Shu et al., 2008; Albuz et al., 2010). The latter dbcAMP-mediated biphasic IVM system, in particular, has been studied extensively and is used commercially in domestic animal ART (Leal et al., 2018).

The landmark study by Conti et al. (Park et al., 2004), demonstrating that the EGF-like peptides are the key inducers of oocyte meiotic maturation in vivo, inspired their use in cAMP-mediated biphasic IVM culture systems. This was based on the important work by Downs and Eppig who showed that EGF could induce meiotic resumption in the presence of a PDE inhibitor (Downs et al., 1986, 1988), introducing the notion of induced-IVM c.f. spontaneous IVM (Gilchrist and Thompson, 2007). Many papers then demonstrated that EGF or EGF-like peptides are the preferred inducers of oocyte meiotic resumption, compared to FSH, in cAMP-mediated biphasic IVM systems (Akaki et al., 2009; Albuz et al., 2010; Richani et al., 2014; Sugimura et al., 2015, 2018; Fig. 2).

An important central theme that emerged from all of these studies of cAMP-mediated IVM in terms of IVM efficacy: namely, that it is not advisable to allow the oocyte to spontaneously resume meiosis when it is removed from the follicle for IVM, that GVBD needs to be actively prevented in the earliest phases of IVM, and that GV-arrest can be overridden with EGF-like peptides inducing oocyte maturation in vitro (Gilchrist, 2011).

CNP-mediated pre-IVM

The seminal discovery in 2010 that CNP (Zhang et al., 2010) is the natural oocyte maturation inhibitor led to a re-evaluation of old knowledge of the role of cGMP (Hubbard and Terranova, 1982) as a more physiological inhibitor in pre-IVM, leading to the development of a CNP-mediated pre-IVM system for bovine oocytes (Franciosi et al., 2014). CNP-mediated pre-IVM was then validated in a range of other species including mice, pigs, sheep, goats and humans (see section ‘Effects of Biphasic IVM on the COC’). A critical component of CNP-mediated pre-IVM is the inclusion of oestradiol, needed to maintain CC expression of the CNP receptor NPR2 (Zhang et al., 2011). Based on the knowledge gained by using the EGF-like peptides in cAMP-mediated biphasic IVM to induce meiotic resumption (see section ‘cAMP-Mediated Pre-IVM’), EGF-like peptides were rapidly adopted in CNP-mediated biphasic IVM (Fig. 2), initially using mouse oocytes (Romero et al., 2016) then using human oocytes (Sanchez et al., 2017). These studies formed the basis for the capacitation-IVM (CAPA-IVM) culture system, currently in clinical use today (Vuong et al., 2020b), which is examined in detail in the sections below.

Pre-IVM duration and FSH-priming

The optimal duration of pre-IVM remains an open question. Initial biphasic IVM studies cultured COCs in pre-IVM ranging from ∼1 h (collection medium; Luciano et al., 1999) to 20 h (Funahashi et al., 1997). The widely used dbcAMP-mediated biphasic IVM system in domestic animals uses a 20–24 h pre-IVM step (Akaki et al., 2009; Appeltant et al., 2015). Using cAMP-mediated pre-IVM (Richani et al., 2014; Li et al., 2016) and CNP-mediated pre-IVM (Zhao et al., 2020) improves oocyte quality and developmental competence in a temporally dependent manner. It has been hypothesized that the length of pre-IVM required is inversely related to the developmental status of the COC at collection, i.e. follicle size and/or extent of FSH-priming (Zhao et al., 2020; Richani and Gilchrist, 2022) (Fig. 3). Support for this hypothesis comes from the following: in unstimulated cows, a 5-day cAMP-mediated complex pre-IVM culture system was effective at inducing developmental competence in oocytes from small antral follicles that otherwise had very low developmental potential (Garcia Barros et al., 2023); in unstimulated mice, 48-h CNP-mediated pre-IVM is effective (Romero et al., 2016); in mildly stimulated mice (23 h post-PMSG), 24 h CNP-mediated pre-IVM culture was most effective at improving oocyte developmental competence (blastocyst rate and hatching rate) compared with shorter pre-IVM durations (Zhao et al., 2020); and in fully stimulated mice (46 h post-PMSG), a 2-h CNP-mediated pre-IVM culture was more effective than 24 h of pre-IVM in terms of blastocyst production and pregnancy rates (Santiquet et al., 2017a). Using human oocytes from small follicles from patients with PCOS stimulated for 3 days with hMG, CNP-mediated pre-IVM periods of 24 and 46 h efficiently supported meiotic arrest, however, no further gain in oocyte quality or blastocyst yield was recorded when the pre-IVM period was extended from 24 to 46 h (Sanchez et al., 2017). Hence, it can be hypothesized (Zhao et al., 2020; Richani and Gilchrist, 2022) that pre-IVM of some duration and complexity is required in zero- or mild-stimulation scenarios, whereas short or no pre-IVM is required where the patient receives extensive FSH-priming (Fig. 3).

Figure 3.

Hypothesized tradeoff between hyperstimulation and IVM culture complexity. Hypothesized relationship between the extent of FSH-priming (dose/duration) of the patient prior to oocyte retrieval and the duration and complexity of the pre-IVM phase needed to generate a developmentally competent oocyte.

Practical considerations of biphasic IVM for the clinic

Patient selection

Like other types of IVM, theoretically, biphasic IVM can be indicated for patients such as women with PCOS, women with high AFC, women with normal ovarian reserve but ovarian resistance to gonadotrophins, and women with estrogen–sensitive cancers. However, to date, biphasic IVM has been reported in just three types of patient, including women diagnosed with PCOS, women with a high AFC, and women with gynaecological malignancies (Vuong et al., 2020a,b; Kirillova et al., 2021). Like all ART, maternal age is the key determinant of treatment success using biphasic protocols, and age considerations are identical between biphasic and any other IVM protocol.

Gonadotrophin priming

Conventionally, IVM is performed without gonadotrophin priming before immature oocyte retrieval, but in practice FSH-priming and/or hCG-triggering are common in most clinical IVM protocols. The recent large RCT, which included one arm of CAPA-IVM, stimulated patients with highly purified hMG 150 IU/day for 2 days (Vuong et al., 2020a). However, there is limited evidence regarding the effect of FSH-priming or the optimal dosage of FSH/hMG used in IVM but a common dosage is FSH 100–150 IU/day for 2 or 3 days (Vuong et al., 2020a,b). A clinical trial is required to clarify these issues. However, an important point is that an hCG-triggered cycle, or a truncated IVF cycle (FSH + hCG), as has been commonly used in many IVM centers, is incompatible with all biphasic protocols, as an hCG trigger makes it impossible to maintain the oocyte meiotically arrested in the pre-IVM phase of a biphasic IVM system. FSH priming would typically start in the early follicular phase but indeed can start on any day of the cycle, including in the luteal phase, or on any day in acyclic women, to best suit the patient’s and clinic’s convenience.

Oocyte retrieval timing and procedure

The oocyte retrieval procedure for biphasic IVM is similar or identical to other IVM protocols. There is a learning curve for the physician to perform an OPU from small antral follicles. Although closed-circuit needle flushing systems have been developed to avoid blood clots in the aspirated follicular fluid, the optimal technique for oocyte collection from small antral follicles requires further study. Up to now, there is no study in humans to identify the optimal timing of OPU for biphasic IVM. The timing of biphasic IVM OPU is currently set at 42–46 h after the last hMG/FSH injection (Vuong et al., 2020a), based on animal models of FSH-coasting prior to IVM, which improves oocyte quality (Nivet et al., 2012). In cattle, the fine tuning of this time interval is based on acute temporal effects of FSH withdrawal on critical molecular aspects of acquisition of oocyte developmental competence, such as RNA processing functions and translation capacity, and mechanisms regulating chromosome segregation (Labrecque et al., 2013). Such detail is lacking in relation to the human IVM protocol and so it is unknown if the currently used interval of 42–46 h is effective, but previously it was shown that there is no difference between 2 and 3 days of coasting in FSH-primed IVM cycles (Mikkelsen et al., 2003).

Fresh or frozen transfer

A randomized, controlled pilot study was designed to compare the effectiveness and safety of a freeze-all strategy with fresh ET in women with a high AFC undergoing biphasic IVM (Vuong et al., 2021). The results suggested that a freeze-all strategy in a biphasic IVM protocol can significantly increase the rate of ongoing pregnancy resulting in live birth compared with fresh ET. Other fertility outcomes and complication rates did not differ between the two groups, although larger studies with longer follow-up are needed to confirm the comparative safety of frozen versus fresh ET in biphasic IVM (Vuong et al., 2021). Hence, consistent with nearly all current FSH-primed standard IVM protocols (Ortega-Hrepich et al., 2013; Walls et al., 2015), freeze-all is the current clinical practice for the biphasic IVM clinic.

Practical considerations of biphasic IVM for the laboratory

While embryologists performing routine IVF are occasionally confronted with partially expanded COCs (containing GV or Metaphase I oocytes), most embryologists are rarely confronted with, or purposefully search for, fully compacted GV-stage COCs. The compacted COCs used for biphasic IVM are identical in appearance to those used in standard IVM, and in both cases laboratory personnel need to be trained in the processing of bloody follicular aspirates through cell strainers to retrieve compact COCs, which are morphologically distinct from expanded COCs from IVF cycles. An important difference between biphasic IVM and standard IVM protocols is that the COCs require special ‘collection and searching’ conditions to ensure that oocytes are kept under meiotic arrest (e.g. by using CNP in collection medium) in the period before pre-IVM culture commences. During the COC ‘collection and search’ period, that should be kept short and under well-controlled physical conditions of temperature and pH, embryologists must be especially gentle with the COC to carefully preserve its integrity. This is important as pre-IVM systems rely on ensuring adequate CC support of the oocyte.

Biphasic IVM requires two additional days of oocyte culture in the laboratory compared to routine IVF, and one additional day of culture compared to standard IVM. This requires three specialized media: collection/handling medium and pre-IVM medium, both containing a meiosis-inhibiting substance in basal IVM medium, and an oocyte maturation medium. These media are currently not available commercially—rather the specific active reagents (e.g. CNP) are added from laboratory in-house prepared stock concentrates to commercial IVM medium. As discussed in detail in the sections above, there are a number of different methodological approaches to biphasic IVM. In the published CAPA-IVM protocol (Vuong et al., 2020b), the base medium is Medicult IVM media supplemented with 10 mg/ml human serum albumin, 5 ng/ml insulin, and 10 nM oestradiol, which, for the 22–24 h pre-IVM phase, is supplemented with 25 nM CNP and 1 mIU/ml FSH, and thereafter COCs progress to the IVM phase for 30–32 h in base medium with supplements plus 100 ng/ml amphiregulin and 100 mIU/ml FSH (Graphical Abstract). COCs are typically cultured in atmospheric O₂ to support CC metabolism, although the optimal O2 tension to apply to biphasic IVM, or even standard IVM, is still unclear. In mice, 5% versus 20% O₂ in the IVM phase affects blastocyst cell allocation, and placental and fetal weights (Banwell et al., 2007; Preis et al., 2007), indicating profound effects of COC metabolism on oocyte developmental programming. IVM O₂ considerations are compounded by media composition, in particular glucose and FSH concentrations (Hashimoto et al., 2000). Once the oocytes reach Metaphase II (MII) they are treated as per oocytes in a routine IVF/ICSI cycle. IVM and biphasic IVM systems require no additional laboratory equipment not usually present in a routine IVF laboratory. Embryologists need some training to perform IVM and biphasic IVM but the learning curve is considerably less than learning, for example, embryo biopsy, or ICSI.

Effects of biphasic IVM on the COC

COC cellular functions

The rationale for pre-IVM culture is to provide an environment, out with the follicle, that can continue to support growth and development of the oocyte such that it can better sustain subsequent embryo development. Such a goal necessitates a culture system that nurtures the COC. The cAMP- and CNP-mediated pre-IVM systems employed to date elicit profound effects of COC cellular biology. The nature of the impact of biphasic IVM on the COC depends on many variables, such as the developmental state of the COC before pre-IVM, the type of cAMP, or cGMP modulator employed, interactions with co-additives such as FSH, the physical and temporal aspects of the pre-IVM culture system, etc. Some of the major cellular effects of cAMP- and CNP-mediated pre-IVM systems on the COC are summarized in Table 2.

Table 2.

Effect of cAMP- and c-type natriuretic peptide-mediated pre-IVM culture on major cellular functions of the cumulus–oocyte complex.

Cellular processes	cAMP- or CNP-mediated pre-IVM	Selected studies*
Preservation of CC–oocyte communication (TZP, GJC)	cAMP pre-IVM	Luciano et al. (2004, 2011), Thomas et al. (2004a), Albuz et al. (2010), Li et al. (2016), and Sugimura et al. (2018)
	CNP pre-IVM	Franciosi et al. (2014), Romero et al. (2016), Sanchez et al. (2017), Santiquet et al. (2017a), Soares et al. (2017), Soto-Heras et al. (2019a,b), Rouhollahi Varnosfaderani et al. (2020), and Zhao et al. (2020)
Chromatin remodeling	cAMP pre-IVM	Nogueira et al. (2003b), Luciano et al. (2011), Lodde et al. (2013), and Garcia Barros et al. (2023)
	CNP pre-IVM	Romero et al. (2016) and Sanchez et al. (2017)
Increased oocyte growth	cAMP pre-IVM	Garcia Barros et al. (2023)
	CNP pre-IVM	Romero et al. (2016) and Sanchez et al. (2017)
CC proliferation	CNP pre-IVM	Sanchez et al. (2017) and Zhao et al. (2020)
Enhanced glycolysis	cAMP pre-IVM	Zeng et al. (2014)
	CNP pre-IVM	Akin et al. (2021b)
Altered mitochondrial function and oxidative phosphorylation	cAMP pre-IVM	Zeng et al. (2014) and Richani et al. (2019)
	CNP pre-IVM	Santiquet et al. (2017a), Zhang et al. (2017a), Soto-Heras et al. (2019b), and Zhao et al. (2020)
Enhanced antioxidant defence	cAMP pre-IVM	Zeng et al. (2014) and Li et al. (2016)
	CNP pre-IVM	Zhang et al. (2017a), Soto-Heras et al. (2019a), Zhenwei and Xianhua (2019), and Akin et al. (2021a)
Enhanced EGF-peptide signaling	cAMP pre-IVM	Richani et al. (2014) and Sugimura et al. (2015, 2018)
	CNP pre-IVM	Akin et al. (2021a)
Improved meiotic synchrony	cAMP pre-IVM	Funahashi et al. (1997), Nogueira et al. (2003a,b), Thomas et al. (2004a,b); Kim et al. (2008), Shu et al. (2008), Albuz et al. (2010), and Rose et al. (2013)
	CNP pre-IVM	Romero et al. (2016), Sanchez et al. (2019), Vuong et al. (2020b), and Zhao et al. (2020)
Increased MII%	cAMP pre-IVM	Zeng et al. (2014) and Garcia Barros et al. (2023)
	CNP pre-IVM	Zhang et al. (2015a), Romero et al. (2016), Sanchez et al. (2017, 2019), Wei et al. (2017), Vuong et al. (2020b), and Akin et al. (2021a)

*

Human studies in bold.

CC, cumulus cell; CNP, c-type natriuretic peptide; EGF, epidermal growth factor; MII, metaphase II; TZP, trans-zonal process; GJC, gap-junctional communication.

Two fundamental features of pre-IVM culture systems are to prevent precocious GVBD upon removal of the COC from the follicle and to simultaneously protect CC–oocyte communication by preventing loss of the intricate CC–oocyte TZP and GJC networks. There is ample evidence from animal studies that most cAMP-modulators and CNP are highly effective at retaining CC–oocyte GJC (Table 2). This serves to retain the CCs in a non-luteinized state, enabling them to continue to serve their basic role of supporting oocyte development through substrate provision. Crucially, protected CC–oocyte GJC via cAMP- or CNP-mediated pre-IVM facilitates large scale remodeling of chromatin within the oocyte’s GV during the course of pre-IVM, required for the ordered cessation of RNA synthesis in preparation for meiotic resumption (Lodde et al., 2007). In addition, pre-IVM can notably alter COC metabolism, including CC glycolysis, oocyte mitochondrial function and capacity to generate ATP, and production of antioxidants, such as glutathione, much of which is dependent on CC–oocyte communication (Table 2; Li et al., 2016). Pre-IVM culture systems also support development of the CC EGF receptor signaling system, enabling them to become responsive to the EGF-like peptides during IVM (Sugimura et al., 2015).

The net effect of the many cellular processes supported by pre-IVM, as outlined in Table 2, is a system that supports ex vivo growth and development of the oocyte before undergoing in vitro meiotic resumption. Oocyte diameter is a yardstick for an oocyte’s overall developmental state. In CNP-mediated pre-IVM, improvement in developmental competence of oocytes from unstimulated mice is accompanied by an increase in oocyte diameter compared to oocytes obtained immediately after isolation: in that study, the presence of FSH during the 48-h pre-IVM phase was a key factor in stimulating oocyte growth (Romero et al., 2016). A 5-day cAMP-mediated pre-IVM system can effectively grow cow oocytes in vitro from a small under-developed state (Garcia Barros et al., 2023). In human oocytes derived from small antral follicles <6 mm, CNP-mediated pre-IVM may also support oocyte growth during culture (Sanchez et al., 2017). In this report, the average diameters of oocytes that remained arrested at the GV-stage after IVM were measured. GV arrested oocytes that failed to mature following an IVM step were ∼3 µm larger if undergoing a prior CNP-mediated pre-IVM culture, compared to those that were directly subjected to IVM. So far, these data on GV-arrested human oocytes suggests that pre-IVM culture supports the acquisition of oocyte developmental competence, consistent with an increase in oocyte size.

Oocyte maturation

Oocytes used for IVM are commonly collected from a range of follicle sizes of differing atresia status, such that their GV chromatin configurations are in various states of preparedness for GVBD (Mattson and Albertini, 1990; Zuccotti et al., 1998). The consequence of this is that upon spontaneous meiotic resumption in vitro a cohort of oocytes will mature asynchronously making subsequent timing of sperm injection problematic. In fact, the principal objective of the original biphasic IVM publication was to synchronize meiosis of pig oocytes during IVM (Funahashi et al., 1997). Pre-IVM culture systems are effective at preparing the oocyte for timely GVBD, leading to a more rapid and even rate of meiotic maturation in the IVM phase (Thomas et al., 2004b; Kim et al., 2008; Zhao et al., 2020). The net effect of the improved developmental programming in pre-IVM culture is commonly an improvement in oocyte GVBD and MII rates in the IVM phase (Table 2). More mature oocytes from an IVM system increases the probability of improving overall pregnancy rates.

In humans, three small trials have compared the efficacy of CNP-mediated pre-IVM (CAPA-IVM) with standard IVM (Sanchez et al., 2017, 2019; Vuong et al., 2020b). All IVM was performed serum free. In all three studies, CAPA-IVM significantly improved maturation to MII compared to standard IVM (Fig. 4). This improvement is noteworthy as human MII rates using standard IVM have been fixed at ∼50% for decades—ironically the same rate that Robert Edwards reported in 1965 (Edwards, 1965)—whereas CAPA-IVM MII rates are consistently between 60% and 70% (Fig. 3; Sanchez et al., 2017, 2019; Vuong et al., 2020a,b). Oocytes remaining arrested at GV after IVM are also reduced with CAPA-IVM relative to standard IVM (Sanchez et al., 2019), with the higher GVBD and MII rates being consistent with the notion that biphasic IVM ‘induces’ meiosis in the IVM step (Gilchrist et al., 2016). Although the increase in MII rate with CAPA-IVM may seem marginal, on a per-patient basis it has a notable effect on eventual embryo yield, i.e. the attrition rate of oocytes collected through to eventual embryos is less.

Figure 4.

Principal laboratory outcomes from the first human trials where standard IVM was compared to capacitation-IVM. MII, mature metaphase II oocyte/COC collected; GQED3, good quality embryo on day 3/COC collected. CAPA IVM, capacitation IVM; COC, cumulus–oocyte complex; Std IVM, standard IVM. Data from (Sanchez et al., 2017; 2019; Vuong et al., 2020b).

Embryo yield

There is substantial evidence from animal studies gained over the past two to three decades demonstrating that biphasic IVM systems provide benefit to the oocyte in terms of its capacity to support preimplantation embryo development. Until recently, these studies have largely focused on the use of cAMP modulators in the pre-IVM phase—this large body of knowledge is reviewed in detail elsewhere (Gilchrist et al., 2016; Leal et al., 2018). CNP-mediated pre-IVM systems are a more recent development and hence there is less evidence than for cAMP-mediated pre-IVM. Nonetheless, recent studies in different animal models confirm the benefits of CNP-mediated pre-IVM on improved embryo development (Table 3). It is noteworthy that a number of other IVM treatment additives were utilized in some of these studies.

Table 3.

Animal studies using c-type natriuretic peptide-mediated biphasic IVM reporting embryo and/or pregnancy outcomes.

Study	Species	Animal age	Follicle size (mm)	Additional treatment required for improvement	Improved blastocyst %	Improved blastocyst quality*	Pregnancies following ET
Franciosi et al. (2014)	Bovine	4–8 years	2–6			✓	–
Zhang et al. (2015a)	Caprine				✓	✓	–
Romero et al. (2016)	Murine		Early antral	FSH + GDF9	✓		–
Wei et al. (2017)	Murine	8–10 weeks	Small antral		✓		–
Santiquet et al. (2017a)	Murine	5–12 weeks		FSH + E2 + BMP15	✓		Improved implantation
Zhang et al. (2017b)	Porcine		3–4	Similar effects with BNP	✓		–
			4–6		✓
			6–8		×
Santiquet et al. (2017b)	Bovine		2–8		×	×	–
Soares et al. (2017)	Bovine	Adult	2–6	Steroids + FSH		✓	–
Zhang et al. (2017a)	Bovine		2–6		✓	✓	–
Zhang et al. (2018)	Ovine				✓	✓	–
Soto-Heras et al. (2019b)	Bovine	Adult	3–8	+IBMX	✓		–
Soto-Heras et al. (2019a)	Caprine	Pre-pubertal			✓		–
Zhenwei and Xianhua (2019)	Bovine	Adult	3–6	+cysteamine	✓		–
Zhao et al. (2020)	Murine	4 weeks			✓		–
Jia and Wang (2020)	Bovine		<3		×	×	–
			3–8		✓	✓	–
Rouhollahi Varnosfaderani et al. (2020)	Ovine		4–6/2–4	PGE2 or AREG in IVM	✓		–
Assareh et al. (2022)	Caprine			PGE2 or AREG in IVM	✓	✓	n.s.

Improvements (in % metaphase II, % blastocyst and quality) following biphasic IVM are normally compared to IVM derived oocytes or non-pre IVM-treated.

*

Improvement in blastocyst quality in terms of: hatching, cell number, higher inner cell mass/trophectoderm ratio or DNA fragmentation.

GDF9, growth differentiation factor 9; BMP15, bone morphogenetic protein 15; BNP, brain natriuretic peptide; IBMX, 3-isobutyl-1-methyl-xanthine; PGE2, prostaglandin E2; AREG, amphiregulin.

In human oocytes, improvements in embryology outcomes, measured by a significant increase in the number of usable embryos per patient and in embryo quality, was evident when applying CNP-mediated pre-IVM compared to standard IVM. Patients had more good-quality embryos available on Day 3 (Fig 4; Sanchez et al., 2017, 2019; Vuong et al., 2020b) and more good-quality blastocysts (Sanchez et al., 2017) compared to standard IVM groups. Notably, a benefit in developmental outcomes from the biphasic IVM strategy was observed on oocytes from the smallest antral follicles <6 mm compared to those >6 mm (Sanchez et al., 2019), highlighting the importance of the initial developmental status of the COC when applying a biphasic IVM system. The net effect of higher oocyte maturation and embryo development rates was that patients treated with biphasic IVM had on average one additional frozen embryo remaining after the first ET, compared to standard IVM patients (Vuong et al., 2020b).

Clinical outcomes using biphasic IVM

Case numbers and main outcomes from the clinical application of CAPA-IVM are summarized in Table 4. To date there have been seven trails using CAPA-IVM in at least one arm of a trial, with a total of 423 biphasic IVM cycles (Sanchez et al., 2017, 2019; Vuong et al., 2020a,b, 2021; Akin et al., 2021a; Kirillova et al., 2021). Five of these seven trials were performed in the same center (IVFMD, My Duc Hospital, Ho Chi Minh City, Vietnam), which is the only center to date to have published live birth outcomes after any form of biphasic IVM. There have been three pre-clinical safety and efficacy trials generating human embryos without ET; one using IBMX-mediated pre-IVM (Spits et al., 2015) and two using CAPA-IVM (Saenz-de-Juano et al., 2019; Sanchez et al., 2017). Outcomes from specific trials are examined in more detail in the sections below.

Table 4.

Study characteristics and principal clinical outcomes from capacitation-IVM studies to date.

Study	Patient type	Stimulation (mean IU FSH/patient)	Cycles (n)a	COC (n)a	MII (%)	GQED3 (%)	Embryos transferred (mean)	CPR (%)	LBR (%)	Live births (n)a,b
Sanchez et al. (2017)	PCOS	680	15	117	70	43	N/A	N/A	N/A	N/A
Sanchez et al. (2019)	PCOS + high AFC	450	20	305	62	24	N/A	N/A	N/A	N/A
Vuong et al. (2020b)	PCOS + high AFC	379	40	700	64	19	2.0	60	47	19
Vuong et al. (2020a)	PCOS + high AFC	373	268	3806	64	21	1.9	51	35	120
Vuong et al. (2021)	PCOS + high AFC	300	40	732	67	23	2.0	35/70c	20/60c	16
Akin et al. (2021a)	PCOS + high AFC	300	30	555	67/55d	20/15d	1.9	67/43d	47/29d	11
Kirillova et al. (2021)	Gynaecological malignancies	0	10	105e	56	N/A	N/A	N/A	N/A	N/A
Total			423	6320						166

a

Total in study treated with CAPA-IVM.

b

Cumulative live births.

c

Fresh transfer/frozen transfer.

d

With/without amphiregulin in IVM.

e

Ovarian tissue oocytes (OTO). AFC, antral follicle count; COC, cumulus–oocyte complex; MII, metaphase II; GQED3, good quality embryo on Day 3; CPR, clinical pregnancy rate; LBR, live birth rate.

Biphasic IVM compared to standard IVM

The first clinical trial of a biphasic IVM system, CAPA-IVM, commenced in 2016 in Vietnam and led to the first reported birth in July 2017 (Vuong et al., 2020b). In this small pilot RCT, designed to compare CAPA-IVM to routine IVM practice, 80 women with PCO or PCOS were randomized to standard IVM or CAPA-IVM after 2 or 3 days of FSH priming (mean 2.5 days, 375–379 IU FSH/patient) with no trigger. Laboratory outcomes are reported in Fig. 4. The majority of patients had two embryos transferred on Day 3. Clinical pregnancy rate per ET was significantly higher, at 63.2%, in the CAPA-IVM group compared to 38.5% in the standard IVM group (Fig. 5; Vuong et al., 2020b). Live birth rate per ET showed a trend toward a higher rate from CAPA-IVM than standard IVM (50.0% versus 33.3%) but the difference was not statistically significantly. Miscarriage and multiple pregnancy rates were comparable in the two groups.

Figure 5.

Principal clinical outcomes from the first randomized controlled trial comparing standard IVM to capacitation-IVM in patients with PCO and PCOS. CAPA IVM, capacitation IVM; PCO, polycystic ovaries; Std IVM, standard IVM. Data from Vuong et al. (2020b).

Results from this pilot RCT (Vuong et al., 2020b) and the preceding three pre-clinical efficacy and safety studies (Saenz-de-Juano et al., 2016; Sanchez et al., 2017, 2019) validated in humans the principal of biphasic IVM established from animal studies, by showing that CAPA-IVM could improve outcomes for women with PCO and PCOS undergoing ART in terms of significantly increased oocyte maturation rates, with potentially better embryo quality and higher live birth rates. RCTs with larger patient numbers and in diverse populations are needed in order to determine whether live birth rates with biphasic IVM are improved compared with standard IVM.

Biphasic IVM compared to conventional IVF

In a large RCT, a total of 546 women with an indication for ART and a high AFC (more than 24 follicles in both ovaries) were randomized to CAPA-IVM (n = 273) with 2 days of FSH priming and no trigger or conventional ovarian stimulation-IVF (n = 273) (Vuong et al., 2020a). More than 70% of patients were typical cases of PCOS. Live birth rate was the major outcome. Cumulative ongoing pregnancy rates at 12 months after randomization was also calculated. IVF patients received in total 5.5 times more FSH than CAPA-IVM patients (2060 versus 373 IU, respectively) over 8.8 days of treatment compared to 2.3 days for IVM patients (Fig. 6; Vuong et al., 2020a).

Figure 6.

Principal outcomes from the first randomized controlled trial comparing capacitation-IVM to conventional stimulated IVF in women with a high antral follicle count. The trial included 546 ART patients randomized to two arms with live birth rate the main outcome measure. MII, mature metaphase II oocyte/COC collected; GQED3, good quality embryo on day 3/COC collected; CAPA IVM, capacitation IVM; COC, cumulus–oocyte complex. Data from Vuong et al. (2020a).

The oocyte maturation rate after IVM was 64% (Fig. 6; Vuong et al., 2020a) which was consistent with previous CAPA-IVM studies (Sanchez et al., 2019; Vuong et al., 2020b). Yet this percentage of MII was significantly lower using CAPA-IVM compared to conventional IVF (64% versus 79%). Numbers of oocytes retrieved/patient, number of MII oocytes, fertilized oocytes, and freezable embryos/patient, were all significantly lower after CAPA-IVM than conventional IVF (Fig. 6). The number of top-quality embryos was significantly lower in the CAPA-IVM versus IVF group (3.2 versus 7.9; Vuong et al., 2020a).

The results showed that live birth rate after the first frozen transfer did not differ significantly between the CAPA-IVM and conventional IVF groups (35% versus 43%, risk ratio 0.81; Fig. 6). Implantation rates were not significantly different between the two groups; 35% versus. 39%. There were no statistically significant differences in live birth rates between the CAPA-IVM and conventional IVF groups in all patients, and in patient subgroups with (36% versus 41%, respectively) and without PCOS (34% versus 48%, respectively). However, because IVF patients had significantly more Day 3 embryos and therefore had, on average, more frozen embryos than CAPA-IVM patients, the cumulative ongoing pregnancy rates at 12 months after randomization was 44% in the CAPA-IVM group and 63% in the IVF group (Fig. 6; Vuong et al., 2020a).

These findings suggest that in women with a high AFC or PCOS, a biphasic strategy such as the CAPA-IVM protocol is a viable alternative ART to conventional IVF, as it is simpler, more convenient and safer than conventional IVF, as it notably reduces drug use and cycle monitoring and eliminates the risk of OHSS. CAPA-IVM provided fairly good oocyte maturation rates and comparable embryo quality to conventional IVF. However, a lesser number of top-quality embryos per cycle and hence lower cumulative pregnancy rates indicate that CAPA-IVM is, overall, currently less effective than IVF (Vuong et al., 2020a).

Safety of biphasic IVM

Safety for the oocyte and embryo

There has been concern that IVM of oocytes derived from small antral follicles, and in particular biphasic IVM with its extended culture time under meiotic arrest, might induce increased epigenetic risks compared to conventional IVF. As a fundamental epigenetic process during oocyte growth, DNA methylation is established at imprinted germline differentially methylated regions (gDMRs) (Hiura et al., 2006; Lucifero et al., 2004; Obata and Kono, 2002). Moreover, maternal effect products are transcribed and stored in the oocyte, which are necessary for correct maintenance of imprinted DNA methylation after fertilization (reviewed by Hanna and Kelsey (2021)). Mouse studies have shown that superovulation or prolonged in vitro follicle growth do not interfere with imprinting establishment, but rather affect its maintenance during pre-implantation embryo development (Anckaert et al., 2009; Denomme et al., 2011; Saenz-de-Juano et al., 2016). As such, the safety of new IVM protocols should be evaluated in blastocysts before the technology can be safely introduced into the fertility clinic. Hence, imprinted DNA methylation was compared in human Day 5 or Day 6 blastocysts, using data from 20 embryos derived from CAPA-IVM with CNP-mediated meiotic arrest in the pre-IVM step, and 12 embryos from conventional stimulated IVF/ICSI in age-matched patients with PCOS (Fig. 7; Saenz-de-Juano et al., 2019). Low input DNA methylation profiling (post-bisulphite adaptor tagging) of 21 gDMRs revealed no differences between CAPA-IVM and conventional IVF blastocysts in terms of combined gDMR methylation (to increase sequencing coverage and hence robustness of comparison) or in terms of individual gDMR methylation in individual blastocysts. Furthermore, RNA sequencing performed on an additional set of 20 blastocysts showed no differences in transcript levels of 21 imprinted genes and of genes involved in methylation maintenance (TRIM28, ZFP7, ZNF445), demethylation (TET1, TET2, TET3), and remethylation (DNMTI, DNMT3A, DNMT3B, DNMT3L, UHRF1) (Saenz-de-Juano et al., 2019).

Figure 7.

Illustration of the design and outcome measures of the two studies comparing the genetics of human embryos generated from biphasic IVM treated oocytes compared to stimulated IVF oocytes. OPU, oocyte pick-up; aCGH, array comparative genomic hybridization; gDMR, germline differentially methylated regions; RNAseq, RNA sequencing. Illustration by Dr Nazli Akin.

To assess the cytogenetic risk, Spits et al. (2015) assessed 18 normal developing cleavage-stage embryos derived from a biphasic IVM system with IBMX-mediated meiotic arrest in the pre-IVM step in patients with PCOS. Individual blastomeres were analyzed using array comparative genomic hybridization and compared with previously published data on normally developing Day 3 or Day 4 embryos (Fig. 7) obtained from conventional stimulated IVF/ICSI cycles (Mertzanidou et al., 2013a,b). It was reassuring that only one out of 18 IVM embryos showed a meiotic abnormality. Although statistical analysis was not possible in view of the low number of embryos assessed, the rates and types of chromosomal abnormalities were similar in IVM and IVF embryos (Spits et al., 2015).

These safety studies comparing embryos derived from biphasic IVM and conventional stimulated IVF were thus reassuring in terms of similar gDMR methylation rates, imprinted gene expression, expression of epigenetic regulators, and chromosome constitution. These pre-clinical safety studies paved the way for biphasic IVM methodologies (CAPA-IVM specifically) to move to the next phase of clinical trials in infertile patients.

Safety for mothers and children

Biphasic IVM has principally been applied to women who are at higher risk or are more difficult to treat with conventional IVF, in particular women with PCOS or a high AFC who are at increased risk of exaggerated ovarian response, OHSS, ovarian torsion, and the related risks of a very high concentration of steroid hormones after ovarian hyperstimulation.

No cases of OHSS were recorded in any of the five recent trials (Table 4) using biphasic IVM comprising a total of 423 biphasic IVM cycles, using 2–3 days of FSH priming before oocyte retrieval (Akin et al., 2021a; Sanchez et al., 2019; Vuong et al., 2020a,b, 2021). In fact, OHSS has never been recorded in any IVM studies using a range of different IVM protocols. In a large RCT comparing CAPA-IVM and conventional IVF cycles (Vuong et al., 2020a) for women with PCOS or high AFC, OHSS was not observed in the biphasic IVM group but occurred in two women (0.7%) with early moderate/severe grade OHSS in the conventional IVF group. Hence, the risk of OHSS is completely eliminated with the use of IVM, including biphasic IVM, even in women with the most severe form of PCOS.

In a small RCT comparing biphasic IVM and standard IVM that followed to live birth (Vuong et al., 2020b), clinical pregnancy, live birth, miscarriage, ectopic pregnancy, preterm delivery were all similar between the two study groups. The subsequent larger RCT comparing CAPA-IVM to conventional ovarian stimulation with IVF, that tracked patients to live birth, also provides reassurance for the safety of biphasic IVM (Vuong et al., 2020a). There were no significant differences between the two groups with respect to the occurrence of pregnancy complications, obstetric and perinatal complications, preterm delivery, birthweight, and neonatal complications (Vuong et al., 2020a).

Interestingly, the 2-year follow-up of children born to participants of the previous RCT of biphasic IVM versus IVF (Vuong et al., 2020a) showed that overall development up to 24 months of age was comparable in those born after IVM or IVF (Vuong et al., 2022). The authors prospectively used the screening tools ASQ-3 and Red Flags questionnaire (Squires et al., 2009) at three time points (ages 6, 12, and 24 months) to assess development of the children. There were no significant differences in ASQ-3 scores between the IVM and IVF groups in the overall study population, and all mean scores were within the normal range at all time points. Children in both groups showed normal growth with respect to body weight over the first 24 months (Fig. 8; Vuong et al., 2022).

Figure 8.

Growth curves (body weight) for children born after biphasic IVM or IVF. Females (A) and males (B) born after biphasic IVM (red line) or IVF (blue line) are plotted against the Centers for Disease Control and Prevention mean, 25th and 75th percentile curves (dotted black lines). Reprinted with permissions from Vuong et al. (2022).

In a recent prospective cohort study in Vietnam, children born after biphasic IVM were propensity score-matched with those born after natural conception and followed up to a maximum of 24 months (Nguyen et al., 2022). The mean age of children at the end of follow-up was 15 months. The proportions of babies with any abnormal ASQ-3 score or with a developmental red flag were not statistically different between children from the biphasic IVM group and babies conceived naturally (Nguyen et al., 2022). This study further supports the lack of any negative effect of biphasic IVM on childhood physical and mental development.

Applications of biphasic IVM in fertility preservation

Fertility preservation in female cancer patients is an important emerging indication for IVM. Although random start protocols have made ovarian stimulation a feasible strategy in a large group of cancer patients, a 2-week delay of the start of cancer treatment is not possible in a subset of cancer patients whose medical condition requires urgent chemotherapy, such as in some patients with Hodgkin lymphoma or leukemia. In these patients, IVM of immature oocytes retrieved transvaginally followed by vitrification of mature oocytes is an option, at least if they have a good functional ovarian reserve (Sonigo et al., 2016). Cancer patients who prefer to start chemotherapy as soon as possible, who decline ovarian tissue cryopreservation (OTC) or who are medically unfit to undergo abdominal surgery for OTC, are also suitable candidates for this approach, although evidence of the reproductive potential of vitrified IVM oocytes is limited, with only one live birth reported to date (Grynberg et al., 2020). Transvaginal egg retrieval for IVM in the context of fertility preservation does not require hormonal pretreatment and can be combined with OTC from the contralateral ovary during a single intervention to maximize the amount of cryopreserved material in selected cancer patients (Delattre et al., 2020). The use of biphasic IVM protocols has not yet been reported in the context of oocytes collected transvaginally from cancer patients.

Upon surgical excision of ovarian tissue and processing in the laboratory, COCs can be easily aspirated from visible antral follicles in the excised ovarian tissue and are also released from follicles that are ruptured during the OTC process. IVM of immature oocytes harvested as such from extracorporeal ovarian tissue is referred to as ovarian tissue-derived oocyte IVM (OTO-IVM) or ex vivo IVM, and is considered experimental in view of the low number of reported live births from this technique (Segers et al., 2020) and the paucity of human safety data, although this is a routine and commercial procedure in domestic animal breeding. As OTO-IVM can usually be offered in most cases where OTC is planned, this technique holds considerable potential for cancer patients, although there are limitations. First, the availability of COCs from excised ovarian tissue depends on the surgical technique used; high numbers of COCs can be obtained when bi- or uni-lateral oophorectomy is performed, and a complete ovary is decorticated. The COCs are released upon dissecting the cortex from the medulla of the ovary, as small antral follicles are located in the junctional zone between the cortex and medulla. However, surgical excision of small pieces of the surface of the ovary will inevitably yield a very limited number of COCs (Segers et al., 2020). Second, OTC is performed in laboratories with specific expertise, which may require transportation of the tissue if the surgical procedure is performed at a different location. Current OTC transportation protocols at 4°C may be incompatible with OTO-IVM requirements (Nikiforov et al., 2020).

Data on biphasic IVM in the context of OTO-IVM are scarce and have so far been derived from just one study in Russia (Kirillova et al., 2021). In a series of 10 patients who underwent unilateral oophorectomy owing to gynaecological tumours, ovaries were divided in two equal halves and the COCs from these hemi-ovaries were subjected to standard IVM or CAPA-IVM. CAPA-IVM yielded a significantly higher MII rate than standard IVM (56% versus 35%, respectively) and the only embryos that developed to blastocysts (Kirillova et al., 2021). This provides first proof-of-principal using human material, based on decades-old knowledge from animal studies, that COCs collected directly from ovarian tissue ex vivo is ideal material for a wide variety of biphasic IVM strategies (reviewed; Gilchrist et al., 2016).

Discussion

New directions in biphasic IVM

As CNP-mediated IVM has been a relatively new development, first reported in animal studies in 2014 (Franciosi et al., 2014), the current version of biphasic IVM in clinical practice (CAPA-IVM; Vuong et al., 2020a) can safely be regarded as ‘version 1.0’ only. No doubt there are significant opportunities to improve on this protocol as there has in fact been little detailed animal research (Table 3) and only limited human research to date (Table 4) on various iterations of CNP-mediated pre-IVM culture systems. A logical extension of the existing protocol would be to extend the duration of the pre-IVM phase to several days in order to capture the reproductive potential of the large pool of very small antral follicles (e.g. <4 mm in human) or even from the preantral pool. This would require an improvement on existing oocyte retrieval capabilities, but pre-IVM can already be applied to OTO-IVM oocytes from such small follicles. A recent publication by Garcia Barros et al. (2023) took bovine oocytes from small, very underdeveloped follicles and conducted an extended 5-day pre-IVM, leading to an improvement in oocyte developmental competence.

New directions in pre-IVM are likely to include development of improved IVM base media formulations including modern antioxidants, examination of alternate, or additional meiotic inhibitors, such as other natriuretic peptides (Zhang et al., 2005, 2015b), alternate meiotic inducers in the IVM phase such as neuregulin (Dellaqua et al., 2023), co-addition of other growth factors such as GDF9, BMP15, cumulin (Akin et al., 2022; Hussein et al., 2006; Mottershead et al., 2015; Sudiman et al., 2014; Sugimura et al., 2014), or growth factor cocktails that have shown promise in animals such as fibroblast growth factor 2, leukemia inhibitory factor, and insulin-like growth factor 1 (Yuan et al., 2017). Further improving the physical culture microenvironment of the COC for longer culture by using microfluidics or culture matrices, and the fine-tuning of the metabolic needs of the growing complexes, might add to the viability and longer-term pre-IVM culture of human oocytes.

Role of biphasic IVM in futuristic ART: in vitro follicle growth and in vitro gametogenesis

Standard IVM is not well suited to oocytes from small antral follicles whereas, in contrast, biphasic IVM is effective for less developmentally competent oocytes. Hence, biphasic IVM is likely to be important to futuristic ART such as in vitro follicle development (IVD) and in vitro gametogenesis (IVG). IVD has proven very challenging in higher mammals, including humans, with a maximum follicle size grown in vitro to date of ∼0.5–1 mm (McLaughlin et al., 2018; Xiao et al., 2015; Xu et al., 2021). Human oocytes from such small follicles have very low meiotic competence and no developmental competence, making triggering them with hCG (Xu et al., 2021) and even standard IVM entirely unsuitable as next steps to attain MII oocytes, thereby mandating the need for an advanced biphasic IVM strategy to capture their reproductive potential. An advanced biphasic IVM strategy is even more important in IVG, where oocytes are generated from stem cells (Hikabe et al., 2016). Biphasic IVM may be valuable in these scenarios as it shows potential to endow developmental competence in vitro on oocytes that are otherwise developmentally incompetent. This suggests that advanced future iterations of biphasic IVM are likely to become integral to sophisticated ART such as IVD and IVG.

Future clinical indications for biphasic IVM

Given the effectiveness, safety, and patient convenience and compliance, as well as the good progress in improving biphasic IVM culture systems, it is worth considering whether this newly developed ART could open additional clinical opportunities in the future. Table 5 illustrates the current common and less common indications for biphasic IVM and how these may change with time and with further IVM improvements. There is significant new interest in the application of IVM to oncofertility; for both OTO and OPU collected oocytes (Ataman et al., 2022), and we can expect a significant uptake of biphasic IVM methodologies for this scenario. Owing to its simplicity (minimal stimulation and monitoring) and efficacy, and the fact that IVM patients can travel with certainty very soon after oocyte retrieval as there is no risk of OHSS (Walls and Hart, 2018), in the future we could think of applying biphasic IVM for cross-border fertility treatment, and to treat infertile couples in the context of pandemics such as coronavirus disease 2019 and others. Similarly, it is not unlikely that biphasic IVM may be an attractive approach for selected women who turn to elective oocyte cryopreservation to postpone motherhood, more specifically for those with a good functional ovarian reserve who prefer a mild-approach procedure without hormonal side effects and with minimal disruption of their day-to-day activities.

Table 5.

Current and potential future clinical indications for biphasic IVM.

Indication	Current indication	Potential future indication
PCOS	Common	Common
Women with high AFC	Uncommon	Common
Women with normal AFC	Rare	Uncommon
Women with low AFC	Rare	Uncommon
Women with ovarian resistance to gonadotrophins	Uncommon	Uncommon
Onco-fertility—from OTO	Uncommon	Common
Onco-fertility—from oocyte retrieval	Uncommon	Common
Social egg freezing	Rare	Common
Cross-border fertility treatment	Rare	Common

AFC, antral follicle count; ROS, resistant ovary syndrome; OTO, oocyte–tissue oocyte.

Biphasic IVM is a step towards hormone-free ART

The recent CAPA-IVM clinical trials typically used a total of 300 IU of FSH/patient over just 2.0–2.3 days of stimulation, yielding clinical pregnancy rates of ∼50% (Fig. 6 and Table 4; Vuong et al., 2020a). Hence, it is only a small step from this protocol to removing gonadotrophin treatment entirely. Hormone-free ART would bring enormous benefits to the patient as it would prevent side-effects and discomfort, and eliminate the need for frequent hormone and ultrasound monitoring, which together would bring significant cost savings. Finally, it would greatly simplify infertility treatment as oocyte retrievals could be scheduled at short notice, at any stage of the cycle, and at the patient’s and clinic’s convenience. Such an advance could be transformative for reproductive medicine. The current major barrier to the use of biphasic IVM to treat all infertile patients is the narrow range of patient types from which sufficient COCs can be collected to make IVM a viable treatment alternative to conventional IVF. Using current best-practice biphasic IVM in patients with a high AFC, only half the number of good quality embryos are available for freezing compared to conventional IVF (Vuong et al., 2020b). Hence, clear goals for IVM are to improve the yield of COCs collected per follicle aspirated in an IVM oocyte retrieval, and to further improve IVM culture systems to enhance oocyte developmental competence and equalise cumulative pregnancy rates with conventional IVF. It is only through such advances will we make progress towards hormone-free ART for the wider infertile population.

Conclusion

This review has described how recent advances in basic oocyte biology and their application in animal models has shown the way to increase the clinical efficacy of IVM in human. Over the last 5 years, advances in clinical IVM have been supported by large, well designed RCTs performed in centers with expertise in both IVM and in standard IVF, showing equivalent pregnancy and live birth rates but lower cumulative live birth rates owing to a lower number of cryopreserved embryos obtained after IVM. Hence, IVM can now be confidentially added to the list of clinically validated treatments that should be available to reproductive medicine specialists working in leading centers worldwide.

It is fallacious to suggest that IVM should replace IVF. Currently, IVM is only applicable in certain groups of patients, particularly those with PCOS, cancer patients for fertility preservation before chemotherapy, and women who are gonadotrophin resistant. It has wider application for women who wish to avoid the potential and real side effects of superovulation with exogenous FSH, allowing minimal or no ovarian stimulation before egg collection. Instead, IVM should be one more tool available for the optimum treatment of these groups of patients, running alongside a conventional IVF program. Although additional laboratory and clinical skills are needed, they are not difficult to acquire, and centers with a high-quality IVF program will quickly obtain pregnancies after IVM.

More progress is needed. There are over 1 million IVF births per year whereas studied IVM pregnancies and babies still number in the low thousands. There will be further technological refinements in IVM and further improvements in blastocyst and pregnancy rates, and in time an equivalence of cumulative live birth rates with conventional IVF. Progress in the science of ART has been unrelenting over the past three decades and IVM is the logical ‘next step’ for many patients who require ART.

Data availability

No new data were generated or analyzed in support of this research.

Acknowledgements

The authors are grateful for Drs Emily Frost and Nazli Akin for preparing the illustrations in Figs 1 and 7, respectively. The authors gratefully acknowledge the careful preparation of the CAPA-IVM research components for the preclinical and pilot studies by Mrs Heidi Van Ranst.

Authors’ roles

R.B.G. and J.S. conceived the idea of the study and wrote the study proposal. All authors wrote individual section(s), which were then edited by R.B.G. and J.S. Tables were prepared by R.B.G., F.S., S.R., and L.N.V. Figures were conceived and prepared by RBG, except Fig. 7 by E.A. and Fig. 8 by T.M.H. and L.N.V. All sections of the manuscript were reviewed by all authors.

Funding

R.B.G.’s work on IVM is funded by an Investigator Fellowship (APP2009940) from the National Health and Medical Research Council of Australia, and by Open Philanthropy. L.N.V. and T.M.H.’s work on IVM protocol has been funded by the Vietnam National Foundation for Science and Technology Development (NAFOSTED) under grant number FWO.108-2022.01. J.S.’s work on Capacitation IVM was funded by grants from Fonds voor Wetenschappelijk Onderzoek Vlaanderen (FWO) (Nr G.0343.13), Excellence of Science (EOS) FWO-FNRS (National Fund for Scientific Research) grant (Nr GOF3118N), FWO/Nafosted (Nr GOD97.18N), and by the Foundation Against Cancer (Hope Project File C69).

Conflict of interest

R.B.G. is a consultant to City Fertility CHA Global on IVM technologies and is a Scientific Advisory Board member to City Fertility CHA Global and CooperSurgical, who sell IVM products. R.B.G. has received travel support from Cooper Surgical. F.S. and S.R. hold a patent on IVM and have received funding from Lavima Fertility Inc, a spinoff company from Vrije Universiteit Brussel conducting R&D in IVM, to support clinical research in IVM. M.D.V. is an employee of UZ Brussel and Vrije Universiteit Brussel which have a license agreement with Lavima Fertility Inc., and is a Scientific Advisory Board member to CooperSurgical. MDV has also received speakers’ fees from Merck, Gedeon Richter, IBSA, Ferring, & Cooper Surgical, and participates as a Scientific Advisory Board member on Cooper Surgical. L.N.V. and T.M.H. have received funding from Lavima Fertility Inc to support clinical research on IVM, and received honoraria from Merck, Merck Sharp and Dohme, & Ferring. TMH is on the Scientific Advisory Board to CooperSurgical. JS holds patents on IVM and is CSO and shareholder of Lavima Fertility Inc.