A major obstacle contributing to the decreased efficacy of in vitro fertilization (IVF) programs is the prevalence of diminished ovarian reserve (DOR) and premature ovarian insufficiency (POI, or premature ovarian failure—POF) in infertile patients. In addition, the age-related decline in follicle number leads to a lower availability of oocytes in IVF cycles in the poor prognosis group characterized by a poor response to ovarian stimulation, resulting in a higher dropout rate of ART cycles and a lower number of live births.
1. Introduction
In a recent report published by the World Health Organization (WHO) in 2022, the global prevalence of infertility is estimated at 17.5%, with the highest rates in the Western Pacific region, reaching a peak of 23.2%
[1]. Despite significant advances in assisted reproductive technology (ART), achieving successful pregnancy rates remains a challenge. The success rate is only around 37% per cycle, and this rate continues to decrease as the mother ages
[2]. In an effort to improve cycle success, healthcare providers often offer additional technologies to standard procedures: for example, the use of stem cell therapy in the treatment of Asherman syndrome and the use of platelet-rich plasma (PRP) or PRP in combination with minimally manipulated endometrial cells to treat a refractory thin endometrium
[3][4][5][6][7].
A major obstacle contributing to the decreased efficacy of in vitro fertilization (IVF) programs is the prevalence of diminished ovarian reserve (DOR) and premature ovarian insufficiency (POI, or premature ovarian failure—POF) in infertile patients. In addition, the age-related decline in follicle number leads to a lower availability of oocytes in IVF cycles
[8] in the poor prognosis group characterized by a poor response to ovarian stimulation, resulting in a higher dropout rate of ART cycles and a lower number of live births. The major limitation to the effectiveness of IVF programs is the limited ability to significantly influence the number and competence of oocyte–cumulus complexes obtained.
POI is a multifaceted condition affecting 1% of women of a reproductive age. It is characterized by elevated gonadotropin levels, including follicle-stimulating hormone (FSH), and irregular or absent menstruation before the age of 40
[9]. Another condition, DOR, also known as a poor ovarian response to stimulation, is also quite heterogeneous in nature and is commonly seen in fertility treatments in regularly menstruating women
[10]. It is worth noting that many cases of DOR, observed in clinical practice, are related to physiological factors and represent an age-related decline in ovarian reserve.
Clinical guidelines have been established worldwide for predicting the ovarian response to stimulation and diagnosing a decline in ovarian reserve. According to the clinical guidelines of the European Society of Human Reproduction and Embryology (ESHRE), ovarian reserve is preferably assessed using indicators such as the number of antral follicles or the level of anti-Muellerian hormone (AMH). According to the Bologna criteria, an AMH level below 1.1 ng/mL is an indication that the patient may have a potential poor ovarian response
[11]. Meanwhile, the American College of Obstetricians and Gynecologists and the American Society of Reproductive Medicine suggest that an AMH level below 1 ng/mL and an FSH level above 10 mIU/mL indicate DOR
[12][13][14].
To improve the results of ovarian stimulation, researchers have explored a number of strategies. These include switching to higher doses of gonadotropins, altering ovarian stimulation protocols, using “freeze-all” strategies and various other approaches
[15]. However, these approaches raise concerns about potential adverse effects on oocyte quality and the differential impact of different protocols on ART outcomes in different cycles
[16]. The proposed methods for ovarian stimulation in patients with DOR include a range of strategies, including delayed-onset gonadotropin-releasing hormone (GnRH) agonists
[17][18], microdose GnRH agonist
[19][20][21], multiple-dose GnRH antagonist
[22][23][24], mild ovarian stimulation
[25], long GnRH agonist
[26][27][28], GnRH antagonist
[29][30][31], GnRH antagonist/letrozole
[32][33], progestin-primed ovarian stimulation
[34][35], short GnRH agonist
[36][37], duo-stim and random-start ovarian stimulation protocols
[11]. According to Di et al.
[38], among these treatment regimens, the delayed-start GnRH agonist and the microdosed GnRH agonist have emerged as the two most promising approaches for the treatment of DOR and show favorable clinical outcomes.
It is evident that a change in gonadotropin regimens, where drug administration does not begin until day 2 of the menstrual cycle, when follicles have entered the gonadotropin-dependent growth phase, affects a pre-programmed and established follicular pool
[39]. The fate of these follicles is determined 90–120 days before the start of the IVF protocol, emphasizing the crucial role of the gonadotropin-independent phase of follicular development in shaping the fate of future oocytes and embryos
[40]. Therefore, there are compelling reasons to explore strategies that target primordial follicles before they enter the gonadotropin-dependent stage of development
[41][42][43][44].
The prospect of reactivating dormant follicles in the ovarian cortex has attracted considerable attention after Li et al. demonstrated their reactivation and successful development into healthy offspring using an in vitro activation (IVA) (protocol
[45]). In this innovative approach, PTEN (phosphatase and tensin homolog) inhibitors and PI3K (phosphatidylinositol 3-kinase) stimulators, in particular AKT (serine/threonine protein kinase 1) stimulators, were used. The work of Li et al. is a fundamental achievement in ovarian reactivation techniques. In addition, studies of ovarian biopsies from women with POI have shown that nearly 40% of these women retained some (30%) or many (9%) follicles in the ovarian cortex
[46].
Although mammalian ovaries harbor numerous follicles, the majority of them remain dormant for long periods of time, even decades. In 2004, Tilly’s research challenged the conventional assumption of a limited number of oocytes in female mammals by identifying ovarian stem cells in the ovarian cortex
[47]. Subsequent studies in mice and other mammals, including humans, confirmed these findings
[43][44][48][49][50][51][52]. Nevertheless, the data from these studies have been repeatedly questioned
[53]. Currently, two types of ovarian stem cells have been identified: pluripotent cells with a diameter of 2–4 μm, called Very Small Embryonic-Like Stem Cells (VSELs), and the ovarian stem cells themselves, which have a diameter of 5–8 μm and express the cytoplasmic isoform of the transcription factor Oct-4B. Among the markers, there are both common and typical markers of stem cells, such as Oct-4, Nanog and SOX-2, and more specific ones, such as STRO-1, FRAGILIS, LIN28, etc. It is likely that stimulation of these cells is one of the possible mechanisms underlying such a method of activating the ovarian reserve as stem cell transplantation. It is unclear how exactly mesenchymal stem cells improve ovarian function. Extracellular vesicles (EVs), a critical component in the new paradigm of intercellular communication, enable transplanted stem cells to communicate with cellular components of ovarian tissue and secrete pro-regenerative substances.
2. Premature Ovarian Insufficiency
The ovarian reserve represents the cumulative number of follicles in the ovaries. It includes both nongrowing follicles and follicles that are recruited at various stages of growth, from preantral to antral, to ovulation. According to the widely accepted theory, women are endowed with a limited number of ovarian follicles at birth. This number declines markedly during intrauterine development, dropping from a peak of about 7 million to 1 million at the time of birth. This decline continues through childhood, with about 400,000 follicles remaining at the onset of menarche. Eventually, as menopause approaches, fewer than 1000 follicles remain in the ovaries
[54][55].
Furthermore, this gradual decline in the number of follicles with age is accompanied by a sequence of reproductive events. This sequence begins with a decline in fertility, progresses to the natural onset of infertility and often includes disruptions of the menstrual cycle that eventually lead to the complete absence of menstruation at menopause. Theoretically, these phases develop over “fixed time intervals” that precede the transition to the next phase
[56][57]. Consequently, three different scenarios may manifest: a “normal” age-related decline in ovarian reserve, a DOR, due to prenatal factors, or a decline in ovarian reserve resulting from the influence of adverse environmental factors or nutritional deficiencies
[58].
The etiology of an early loss of ovarian reserve is associated with several factors, including iatrogenic influences, autoimmune factors, chromosome X-associated abnormalities and point gene mutations
[48]. In particular, iatrogenic causes such as ovarian surgery, gonadotoxic therapies and certain medical treatments contribute significantly to the development of POI and decreased ovarian reserve
[48]. Of the various iatrogenic factors contributing to POI, a significant proportion is attributed to ovarian surgery, which accounts for approximately 64% (excluding bilateral ovariectomy)
[48]. In addition, radiotherapy and chemotherapy used in the treatment of a variety of diseases, including both malignant and benign conditions, are also common causes of POI
[48][59]. Mutations and decreased expression of critical genes related to DNA repair and mitochondrial function may accelerate the depletion of follicular reserve in POI and middle-aged women, respectively. Whereas the process of follicle formation was previously thought to be predominantly under the direct control of the central nervous system, current understanding emphasizes the essential role that paracrine mechanisms play in its regulation.
The decrease in ovarian follicular reserve is due to the periodic, sequential activation of primordial follicles coming out of dormancy. Primordial follicle activation is a crucial step in folliculogenesis that ultimately culminates in the selection of a single oocyte for ovulation. An abnormal acceleration of this activation process can significantly reduce the ovarian reserve
[60]. Currently, depletion of POI is thought to be caused by excessive acceleration of activation of the pool of primordial follicles. Maintaining a high ovarian reserve depends on a balance between activating and inhibitory factors. Recent studies have focused on PTEN/PI3K/AKT/Forkhead Box O3 (FOXO3) and the Hippo signaling pathway
[61], which will be discussed in more detail below.
Modern strategies for controlled ovarian stimulation primarily target antral follicle growth, whereas dormant primordial follicles do not respond to conventional stimulation protocols. Understanding the biological basis of primordial follicle activation is extremely important but has been relatively elusive. Primordial follicle activation is a complex interplay of various local factors and intracellular signaling pathways. Activators such as BMP4/7, GDF-9, KIT ligand, FGF2/7, insulin, GREM1/2 and LIF and suppressors including AMH, LHX8, PTEN, Tsc1m/TORC1, FOXO3a, YAP/Hippo signaling pathway and FOXL2 have been associated with primordial follicle activation
[62][63]. Most of the knowledge about signaling networks and molecules involved in primordial follicle activation has been obtained from rodent premature ovarian failure models
[64][65].
Although the complete mechanism of primordial follicle activation is not yet fully understood, studies conducted on murine models have shown that specific deletion of the PTEN and FOXO3 genes in oocytes promotes the activation and growth of all primordial follicles
[66][67]. The PTEN gene encodes a phosphatase enzyme that negatively regulates the PI3K/AKT/FOXO3a signaling cascade
[68][69]. Stimulation of dormant primordial follicles has also been achieved with the use of PTEN inhibitors and AKT activators in both mice and humans. In the ovaries, the AKT protein plays an important role within the PI3K/AKT/mTOR signaling pathway and is expressed by both oocytes and granulosa cells of human follicles
[61][69]. AKT targets a broad spectrum of molecules that directly and indirectly affect follicle activation
[68][70]. This signaling pathway includes a variety of regulatory molecules that exert inhibitory control within the kinase cascade. Ultimately, this cascade counteracts the transcriptional coactivator Yes-associated protein (YAP) together with its PDZ-binding motif (TAZ), thereby suppressing cell growth
[61]. Maintaining the balance between cell proliferation and apoptosis is necessary to maintain organ size and tissue homeostasis throughout life. In mammals, both processes are coordinated via the Salvador–Warts–Hippo signaling pathway.
Numerous genetic factors have been associated with POI, with certain genes such as GDF-9 (growth differentiation factor 9), BMP-15 (bone morphogenic protein 15)
[49], FOXL2
[50], FSHR (follicle-stimulating hormone receptor)
[51], STAG3 (stromal tumor antigen 3)
[52], XRCC2 (X-ray repair cross-compliment 2)
[53], MCM8
[71], NRIP1, XPO1 and MACF1
[72] showing significant associations. The impairment of primordial follicle activation emerges as a central biological mechanism underlying premature loss of ovarian reserve, regardless of etiology
[73][74]. Animal models, often involving chemotherapeutic agents, mental stress, galactosemia, ovarian antigen peptides, autoimmunity activation (pellucid glycogen ZP3) or genetically manipulated mice, have shed light on the intricate mechanisms behind follicular apoptosis and autophagy inhibition during follicle maturation. However, there may be a number of adverse consequences, such as myelosuppression in the chemotherapy-induced POI or low stability of autoimmunity or mental stress in the POI animal model. The analysis of existing animal models and the POI therapies carried out with them shows that the most effective treatment methods are hormone replacement therapy and stem cell transplantation.
Disturbances in this delicate balance between cell proliferation and apoptosis in the ovaries, as evidenced by the premature activation and rapid depletion of follicles in AMH-knockout mouse models, underscore the importance of ovarian homeostasis in maintaining the follicular pool. It is widely accepted that AMH regulates the amount of growing follicles and influences their selection for ovulation
[75]. AMH primarily exerts a negative regulatory role especially in the early stages of follicular development
[76]. It accomplishes this by suppressing both the recruitment and growth of follicles, effectively dampening the actions of growth factors and gonadotropins, particularly FSH
[77]. Experimental studies have shown that the number of growing follicles decreases by about 40–50% when AMH is introduced into cultured human ovarian cortical tissue
[78]. This highlights the ability of AMH to inhibit the initial stages of follicle development, especially those dependent on FSH. Notably, the absence of AMH leads to a more rapid depletion of the ovarian follicular pool
[79]. In genetically modified female mice with AMH knockout, ovulation ceases at about 16–17 months of age, whereas older wild-type female mice continue to have normal menstrual cycles
[80]. At 13 months of age, the pool of primordial follicles decreases three-fold in female mice lacking AMH compared to wild-type mice. AMH also affects estrogen biosynthesis by inhibiting aromatase activity, resulting in a decrease in estrogen production
[81]. It is noteworthy that AMH concentration in the follicular fluid of patients undergoing in vitro fertilization correlates with estradiol concentration, suggesting a possible role of AMH as a coregulator of steroidogenesis
[75]. In addition, AMH plays an autocrine role in the maturation of normal follicles. Studies in mice have shown that AMH inhibits the first meiotic division of diplotene oocytes
[82]. In human granulosa cells, AMH also shows its influence by blocking proliferation in vitro
[83]. Oocytes within a pool of growing follicles may exert control over a pool of primordial follicles by modulating the expression of AMH
[76]. Age-related alterations in AMH levels contribute to accelerated follicular activation, which in turn contributes to the overall process of ovarian reserve loss
[84].
Numerous preclinical and clinical studies have shown that the field of stem-cell-based therapeutics is very promising and has attracted significant interest due to its potential to treat POI
[85][86]. It is noteworthy that even in cases where the ovaries lose their ability to ovulate, a reserve of dormant follicles remains, offering the possibility of growth under the influence of stem cells
[87][88]. Consequently, further research and development in this field may pave the way for revolutionary therapies in the future.
Results regarding anti-age-associated ovarian hypofunction effects and improvement of ovarian function have been shown after mesenchymal stem cell transplantations in mouse models
[89][90]. One of the first papers to use this method in humans was the study by Herraiz and colleagues
[91], in which autologous ovarian stem cell transplantation (ASCOT) significantly improved ovarian function by increasing the number of antral follicles and oocytes in 81.3% of patients who responded poorly. Although the embryo euploidy rate was low, ASCOT managed to overcome previous limitations and enable pregnancy via enhancement of existing follicles. A number of papers show encouraging results of this technology, which are well described in the review by Fàbregues and colleagues
[92]. Exactly how mesenchymal stem cells improve ovarian function is not entirely clear. In addition to the secretion of pro-regenerative factors, transplanted stem cells may interact with the cellular components of ovarian tissue using EVs.