Restoring Ovarian Fertility and Hormone Function: Recent Advancements, Ongoing Efforts and Future Applications

Abstract

The last 20 years have seen substantial improvements in fertility and hormone preservation and restoration technologies for a growing number of cancer survivors. However, further advancements are required to fill the gaps for those who cannot use current technologies or to improve the efficacy and longevity of current fertility and hormone restoration technologies. Ovarian tissue cryopreservation (OTC) followed by ovarian tissue transplantation (OTT) offers those unable to undergo ovarian stimulation for egg retrieval and cryopreservation an option that restores both fertility and hormone function. However, those with metastatic disease in their ovaries are unable to transplant this tissue. Therefore, new technologies to produce good-quality eggs and restore long-term cyclic ovarian function are being investigated and developed to expand options for a variety of patients. This mini-review describes current and near future technologies including in vitro maturation, in vitro follicle growth and maturation, bioprosthetic ovaries, and stem cell applications in fertility restoration research by their proximity to clinical application.

Advances in cancer therapies have led to substantial gains in survivorship [1]. In particular, the 5-year survival rate for children and adolescents with cancer has risen to above 85% [2]. Despite improved prognoses, life-saving treatments such as chemotherapy and radiation may result in premature gonadal insufficiency, or the inability to produce gametes and gonadal hormones [3]. Consequently, survivorship concerns such as fertility preservation and hormone restoration are increasingly prioritized by patients and physicians [4]. Not all cancer patients or patients who are undergoing chemotherapy or radiation treatments are at increased risk for loss of fertility and endocrine function. However, several medical societies, including those for pediatric patients, recommend that every patient with a new cancer diagnosis be counseled and educated on their risk [4-7]. Alkylating chemotherapies, radiation that targets the brain or abdomen, and stem cell conditioning regimens raises a patient's risk of gonadal insufficiency to “significant” or “high” [3]. Patients at high increased risk of developing premature gonadal insufficiency due to their treatments should be counseled on fertility preservation options.

Current options for fertility preservation are stratified by type of gonads present and pubertal status of the patient. Pubertal patients may use standard assisted reproductive technologies including gamete cryopreservation. However, for patients with ovaries, these techniques can be time intensive and require hormonal stimulation that is incompatible with the treatment plan. Furthermore, gamete cryopreservation is not feasible for prepubertal children who cannot undergo ovarian stimulation and egg retrieval. In contrast to pubertal individuals, whose oocytes are able to undergo meiotic maturation into fertilizable eggs, the immature hypothalamic-pituitary-gonadal axis in prepubertal individuals renders them insensitive to ovarian stimulation with oocytes that are not generally capable of normal fertilization [8].

As the production of euploid eggs occurs along a bell-shaped curve that peaks between ages 21 and 32 years, the quantity and quality of eggs retrieved from adolescent and young adult patients is an important consideration for the development of new and improved reproductive technologies for this clinical population [9]. An option available both to prepubertal and pubertal individuals is ovarian tissue cryopreservation (OTC). OTC involves isolation and cryopreservation of the cortical region of the ovary, which contains primordial follicles that constitute the ovarian reserve, for future autologous retransplantation (ovarian tissue transplantation, OTT) [10]. More than 360 OTTs have been reported, resulting in more than 140 reported live births after OTC [11-13]. However, although cumulatively 95% of participants who undergo OTT have a return to endocrine function post transplantation, the average duration of endocrine function is approximately 2 to 5 years [11, 14-16]. With an estimated life expectancy for a recent cohort of pediatric cancer survivors being between 55.9 and 58.1 years, patients will likely require multiple tissue transplants for long-term hormone restoration [17, 18]. Thus, if OTT is further refined, OTC and subsequent OTT have potential clinical utility that allows for more physiologic hormone restoration when compared to current hormone replacement therapies in addition to fertility preservation. Nevertheless, the safety and efficacy of OTC have generated interest in the expansion of clinical indications to include other contexts in which patients experience premature ovarian insufficiency such as Turner syndrome. Of the 174 ovaries that were cryopreserved using OTC for Turner syndrome patients, 62 (35.2%) of the ovaries contained follicles and thus may be viable for future OTT [19-26].

While OTC and OTT are important breakthroughs from the last 20 years, these techniques have additional limitations. For example, a recent publication of a cohort of pediatric patients who have undergone OTC at an academic medical center indicates that more than 20% received a diagnosis of leukemia or lymphoma [27]. In these patients, OTT may be contraindicated as in vivo xenograft studies using ovarian tissue showed migration of cells with leukemia-specific markers outside the transplanted tissue [28]. Additionally, those with known disease in their ovary are ineligible for transplantation of tissue in its native form due to the unknown risk of reseeding malignancy [2, 29-31]. Therefore, while OTC has enabled patients who could not undergo ovarian stimulation and oocyte retrieval an opportunity to preserve their fertility, additional innovations must be developed to improve the longevity and function of ovarian tissue after transplant, make OTT safe for all patients, and design new options for fertility and hormone restoration. As a result, work to provide expanded access and availability of fertility preservation technologies for oncofertility applications and beyond is an exciting avenue of current scientific research. This mini-review describes current and near future technologies in fertility restoration research by their proximity to clinical application (Fig. 1).

Figure 1.

Fertility preservation and restoration advancements, from the bench to the clinic. Graphical representation of the advancements in fertility restoration from methodologies used currently in the clinic to early preclinical studies. (1) In the clinic: (egg retrieval and freezing; embryo preservation; OTC, ovarian tissue cryopreservation; and IVM, in vitro maturation of cumulus-oocyte-complexes [COCs] collected during OTC). (2) Rounding the mark: optimizing the culture and growth of immature eggs (IVGM, in vitro follicle growth with isolated follicles or in situ within tissue). (3) Ongoing preclinical work: defining the necessary microenvironment needed to engineer a bioprosthetic ovary for humans (ECM, extracellular matrix; soluble factors, paracrine signaling, physical cues, and vascularization).

Ahead of the Pack: In Vitro Maturation

Closest to clinical application is an adaptation of current protocols used for in vitro fertilization (IVF), in vitro maturation (IVM) of cumulus-oocyte-complexes (COCs). COCs are aggregates of an oocyte surrounded by cumulus granulosa cells and can be retrieved from antral follicles following ovarian stimulation [32]. Often COCs collected with immature oocytes can undergo IVM to induce meiosis II in vitro, allowing for cryopreservation of mature, fertilizable eggs. COCs can also be released from antral follicles during ovarian tissue processing for OTC [33]. These recovered COCs may contain oocytes that are developed enough to undergo IVM. A recent systematic review of 12 studies where oocyte cryopreservation was performed with the COCs obtained during ovary tissue processing for OTC found that only 33% of recovered oocytes were able to be matured using current IVM technologies in a mixed population of pediatric and adult patients [34]. While there were some differences in the IVM protocols across centers, this is a reduced rate of maturation compared to adult-only IVF populations, where rates of mature metaphase II (MII) eggs resulting from IVM of recovered oocytes is approximately 70% [35]. These data indicate that age and pubertal status still significantly affect outcomes; however, clinicians and researchers have identified mature oocytes as well as oocytes that have successfully undergone IVM to produce developmentally competent eggs in the OTC processing media both from pediatric and adult patients [36-41]. Therefore, though pediatric patients may continue to face challenges, IVM provides an opportunity for pediatric oncofertility, where options are currently severely limited [37, 38, 42-45].

Many institutions continue to encourage egg cryopreservation over OTC because egg cryopreservation is a standard procedure for IVF clinics. However, oocytes from adolescents and young adults mature at a lower rate than their adult counterparts, and resulting eggs are less likely to be euploid [9, 46, 47]. This raises concerns regarding the future fertilization potential of eggs obtained from very young patients even when IVM is successful. An in vitro technique that can produce good-quality eggs from follicles of earlier stages (primordial, primary, or secondary) within ovarian tissue cryopreserved for future use may provide a controlled environment for growth and maturation and thus deliver a solution for these patient populations.

Rounding the Corner: In Vitro Follicle Growth and Maturation

Techniques to recapitulate folliculogenesis, or follicle development, in vitro are collectively termed in vitro follicle growth and maturation (IVGM) and aim to generate well-developed follicles that can progress to IVM as described earlier. These technologies can be classified into follicle isolation and culture methodologies or in situ methods, where follicles remain embedded in the native ovarian tissue [48-57]. Irrespective of technique, follicle stage is critical to overall outcomes and stage-specific adaptations are required to optimize success rates. Primordial follicles are the initial, quiescent stage of follicles, the most abundant stage of oocytes that exist in prepubertal ovaries and are cryopreserved during OTC, making them a great source of potential eggs for patients. However, primordial follicles are incompletely surrounded by their supporting granulosa cells and are embedded within the dense, growth-prohibitive ovarian cortex. Together, these qualities make isolating intact primordial follicles challenging, requiring a series of mechanical and enzymatic steps [48-50, 58, 59]. This, alongside biologic variability, produces inconsistent results, ranging from 0 to more than 100 isolated follicles per participant [50, 59]. Recent studies in which primordial follicles were isolated and cultured in groups within hydrogel formulations (alginate, collagen, or fibrin) had mixed results regarding viability, with only an approximately 10% increase in follicle size during short-term culture [48, 50-53, 59]. Isolated primordial follicles may activate and begin to progress through folliculogenesis but cannot grow beyond primary follicles, the next major stage of folliculogenesis, without the presence of stromal cells [60]. In 1999, the first study to report the successful culture of primary follicles to early secondary-stage follicles was performed by encapsulating isolated follicles in collagen gels. These follicles, however, were short-lived and demonstrated only 40% viability after 24 hours [55]. Similar results were observed when primary follicles were cultured in alginate [56, 57]. The next follicle stage, secondary follicles, had improved viability in alginate, and a subset were able to develop into antral follicles after 30 days of culture [57]. Development to the antral stage is a key benchmark to allow for COC isolation and IVM, as discussed previously. To address limitations in the development of the earliest follicle stages, strategies have been developed to grow primordial follicles in situ to allow for improved isolation for IVGM.

The first report of human primordial follicles grown in situ, published in 1997, used thin sections of human ovarian cortex cultured on cell inserts and generated viable primary and secondary follicles following 21 days of culture [61]. The authors demonstrated that follicle viability was improved when tissue slices were cultured on an extracellular matrix (ECM)-rich substrate, Matrigel [61]. Further work optimizing the culture of human ovarian cortex slices has used defined culture media including alpha-minimum essential media, follicle-stimulating hormone, human serum albumin, and a mix of insulin-selenium-transferrin [62]. Additional studies in 2014 and 2015 demonstrated that ovarian cortical tissue pieces containing primordial and early transitional follicles can continue folliculogenesis into the antral stage when encapsulated in alginate hydrogels or when supported on ECM-rich “papers” [53, 63, 64]. Together, these results highlight the complexity of early follicle activation and growth in humans and a need for additional studies to generate a robust in vitro system of human follicle culture. Furthermore, these studies demonstrate the ability to identify and isolate viable ovarian follicles from human tissue in a variety of settings, with frozen-thawed tissue generally having no significant differences in follicle viability when compared to fresh tissue [45, 65-68].

Current challenges in IVGM include inefficient and unreliable growth and maturation of secondary follicles into eggs. Published protocols that have used secondary follicles isolated from human cortical tissue with or without in situ growth have resulted in 16 eggs of varying quality from 190 follicles from 84 patients [69-71]. These numbers are much lower than the 10 to 61 eggs required for a woman (aged 34-42 years) to have up to a 75% chance of conception with assisted reproductive technologies using high-quality sperm [72]. Given these limitations in current IVGM methods, a more efficient and reliable strategy to isolate, activate (primordial follicles), grow, and mature follicles to produce high-quality eggs is necessary to maximize fertility preservation options for patients. Effective IVGM for human primordial follicles and beyond likely requires a multistep, stage-specific, and complex culture system to support long-term growth and development into antral follicles that produce competent oocytes for IVM. Finally, while successful IVGM protocols would support the ability to have biological children, they would not address the restoration of ovarian hormone production. Therefore, additional technologies, including the bioprosthetic ovary discussed next, have been envisioned to couple long-term fertility and hormone restoration.

Just off the Block: The Bioprosthetic Ovary—Proof of Concept in Murine Models

Advanced technologies, such as the bioprosthetic ovary, aim to restore complete ovarian function, extend the life of current ovarian tissue transplants, and advance fertility restoration options for individuals with disease in their cryopreserved ovarian tissue. Healthy follicles from a patient would be isolated and autologously transplanted into a defined, engineered environment—ensuring immune compatibility while recapitulating the functional components of the ovary. While currently undefined, considerable efforts are underway to identify optimal biomaterials for the bioprosthetic ovary that will support folliculogenesis and maturation in the clinic. In 2015, transplantation of recellularized ovarian scaffolds in ovariectomized mice allowed for estradiol (E₂) production and follicle growth [73]. In 2017, full restoration of murine ovarian function was achieved when a bioprosthetic ovary, composed of donor follicles and a 3-dimensional–printed gelatin scaffold, was transplanted into an ovariectomized mouse [74]. The pups were born as a result of fertilization following natural ovulation of donor eggs from the bioprosthetic ovary after mating. However, additional research beyond this proof-of-concept study to determine how to increase transplant longevity and translate this bioprosthetic ovary to humans is required. One key condition that would influence the longevity of the transplant is the ability to regulate activation and growth of primordial follicles [75-78]. Recent studies suggest that primordial follicle activation is regulated at least in part by physical properties, and ovarian compartmentalization may provide clues as to how ovarian form supports function [58, 79].

Just off the Block: The Bioprosthetic Ovary—Translational Efforts in Humans (Microenvironment)

Complicating translation of the aforementioned technology into humans is the complexity of the human ovarian microenvironment. The human ovary changes dynamically throughout fetal, postnatal, and pubertal development [79]. The fetal ovary is primarily composed of germ cells with little intervening interstitial cell populations [80]. However, at birth, approximately 40% of ovarian volume is composed of the interstitial component [80]. Pediatric ovaries undergo additional significant changes in size, shape, and subanatomic characteristics, completing compartmentalization into cortex and medulla at the same time that the individual demonstrates secondary sex characteristics [79]. These compartments contain unique interstitial populations, with adult ovaries demonstrating diverse cell populations including granulosa, theca/stroma, smooth muscle/perivascular, endothelial, and immune cells in addition to oocytes [81, 82]. Studies that track interstitial cell populations during murine ovary development identified a subset of cells that are recruited to form the steroid-producing theca layer of late-stage follicles, and functional in vitro studies have revealed that interstitial cell signaling is important for follicle activation, growth, and maturation [83-93]. Cyclical recruitment of ovarian follicles normally occurs under gonadotropin (follicle-stimulating hormone and luteinizing hormone)-dependent mechanisms established during puberty. Pediatric and pubertal patients make up a population that would substantially benefit from new restoration technologies and, as noted earlier, there are significant changes within the ovarian microenvironment in which follicles grow and oocytes mature at this time. Therefore, it is reasonable to conclude that the ovarian microenvironment that includes interstitial cell populations, neighboring follicles, secreted proteins, ECM proteins, and endocrine factors, can affect folliculogenesis and the development of a good-quality egg. Exciting work using in vitro systems to define these changes and their importance in follicle development is ongoing and is critical to determining optimal components of the human bioprosthetic ovary.

Standard in vitro assays used to test the effects of components on growth and maturation of isolated follicles use alginate hydrogel for its ease in gelation, biological inertness, and the ability to modify physical rigidity by toggling the alginate percentage or changing the calcium cross-linker [94]. However, alginate would not be useful in a bioprosthetic ovary intended to support long-term endocrine and ovulatory functions, as it is not easily remodeled by ovarian cells. Additionally, alginate does not support vessel infiltration, which limits key biological signaling that normally occurs through ECM interactions and reduces oxygenation and nutrient exchange of larger transplants.

In contrast to alginate, fibrin is readily degraded by proteolytic enzymes produced by granulosa and theca cells. In vitro, fibrin added to the alginate hydrogel (fibrin-alginate) enhanced the production of meiotically competent oocytes grown from secondary follicles and increased fertilization rates when compared to alginate alone [95, 96]. Fibrin-based materials have also been used to investigate follicle growth and survival in a variety of other species, including caprine, human, and rhesus macaque [97-99]. Isolated rhesus macaque follicles encapsulated in fibrin-alginate produced larger follicles, more E₂, vascular endothelial growth factor (VEGF), and antimüllerian hormone when compared to alginate hydrogels alone [97]. These early studies showcase the ability to improve the production of meiotically competent oocytes in vitro for multiple species by altering the microenvironment of the system.

Further work to improve IVGM has identified other naturally occurring ovarian ECM proteins. For example, a comparison study of alginate and fibrin hydrogels found that murine follicles encapsulated in hyaluronic acid (HA)-alginate produced more eggs and significantly higher levels of E₂ compared to alginate and fibrin hydrogels [100]. Desai et al [101] further supported this finding when murine follicles encapsulated in HA + Matrigel resulted in higher rates of germinal vesicle breakdown, MII formation, and higher E₂ production when compared to controls. Ten years later, the authors reported functional competence when 82% of oocytes retrieved from follicles matured in HA hydrogels were capable of fertilization [102]. Because Matrigel comprises solubilized basement membrane derived from Engelbreth-Holm-Swarm mouse sarcoma making it rich in ECM and growth factors, the improved survival and oocyte quality in these studies can be explained by a variety of reasons [103]. As such, robust clinical application will require delineation of specific factors that contribute to increased follicle survival and oocyte quality.

To further understand the essential environmental cues that regulate folliculogenesis, a map of matrisome proteins and the physical properties of ovaries from model organisms and humans would be of great interest. The matrisome, which includes more than 1000 ECM and associated proteins, can contribute to folliculogenesis with both biochemical and biophysical cues [58, 104-106]. The physical environment is influenced by the density and composition of matrisome proteins. The ovarian cortex, which houses quiescent primordial follicles, is significantly more rigid than the medulla. Recently, a spatial map of porcine ovarian matrisome proteins was generated by proteomics analysis across 2 anatomical planes [107]. Out of the 82 matrisome proteins identified, 42 proteins were significantly differentially expressed across cortical and medullary compartments [107]. These differences in ovarian rigidity are important for folliculogenesis. In mice, treatment of the collagen-dense environment of the ovarian cortex with collagenase and trypsin increased primordial follicle activation in culture [58, 108]. The activation rate was restored by applying hyperbaric pressure, indicating that physical cues can maintain primordial follicle quiescence [58]. However, the reproductive biology field has yet to elucidate whether releasing primordial follicles from their environment alone can increase follicle activation through changes in mechanotransductive cues or whether these properties can be used to control activation [104, 109]. In addition to the physical properties of the matrisome, many proteins contain binding sites or act as ligands that influence transcriptional signals within neighboring cells [105]. Investigating these biochemical properties will be essential for designing a bioengineered environment for use in vitro or in vivo to improve fertility restoration options for patients.

Just off the Block: The Bioprosthetic Ovary—Translational Efforts in Humans (Vascularization)

Vascular networks are essential for follicle growth, nutrient exchange, and peptide hormone transport in vivo. As follicles mature, increasing levels of VEGF are expressed by granulosa and theca cells, stimulating an independent vascular network around the follicle to support its development [110-112]. Therefore, designing materials that promote angiogenesis is essential for the development of a bioprosthetic ovary. Many have reported follicle survival and successful restoration of fertility in vivo using aggregated murine follicles or ovarian grafts encapsulated in fibrin or HA-supported matrices with and without additional growth factors [95, 96, 113-118]. Additional studies of human ovarian tissue that were encapsulated in fibrin clots with the addition of VEGF and xenografted into mice found significantly more proliferating follicles compared to unencapsulated tissue. Vascular structures that penetrated the fibrin-encapsulated tissue expressed both mouse-specific platelet endothelial cell adhesion molecule and human-specific von Willebrand factor, indicating that fibrin encapsulation with VEGF increased revascularization from both donor and recipient endothelial cells, improving follicular growth [119]. Importantly, fibrin glue is already being implemented in some human OTT procedures [120]. In addition to ECM and growth factors, Manavella et al hypothesized that fibrin embedded with proangiogenic adipose tissue–derived stem cells would improve revascularization, limiting hypoxia and ischemic injury to the tissue on transplantation [113-115]. The addition of stem cells in transplants resulted in increased pO₂, total vessel area, primordial follicle survival, and reduced TUNEL-positive follicles [95-98]. These observations, with specific attention to essential microenvironmental factors and revascularization of tissue, are integral to the eventual design of a bioengineered ovary for human ovarian tissue or follicle transplantations.

At the Starting Line: Building Follicles From Scratch

The far future of fertility preservation and restoration may lie in the ability to generate functional ovarian follicles from patients’ own stem cells. Recent reports have described the generation of oocytes and granulosa-like cells to generate functional ovarian follicles from induced pluripotent stem cells (iPSCs) [121]. This work was further extended to the generation of functional oocytes from induced pluripotent stem cells (iPSCs) of XY mice [122]. Additionally, significant efforts are underway to develop hormone-producing granulosa-like cells from human iPSCs, to expand fertility and endocrine restoration options for patients [123]. Although this work is far from a clinical application, it may eventually provide additional fertility preservation and restoration options for a variety of individuals.

Conclusion

With a growing population of patients with premature gonadal insufficiency, there is a need to develop additional options for ovarian fertility and hormone restoration [124, 125]. There have been large advancements in fertility preservation and restoration technologies over the last few decades that are relevant for the practicing clinician. OTC and subsequent OTT is a favorable approach to restore fertility and endocrine function in patients at increased risk for developing premature gonadal insufficiency. However, the effectiveness of these technologies for the pediatric, adolescent, and young adult population must be thoroughly tested, and additional research to support improvements in the functionality and longevity of the ovarian transplant is needed [126, 127]. An overarching goal in the field is to develop new technologies that support fertility and hormone restoration in patients that are unable to use current assisted reproductive technologies, such as those pediatric patients with diseased cells in their ovaries. Development of good-quality human eggs will depend on our understanding of what makes the ideal microenvironment for primordial follicle activation, follicle growth, and oocyte maturation. Elucidating the roles of subanatomical, cellular, extracellular, and secreted factor components of the ovary will improve our understanding of human reproduction and support new ways for restoring function across diverse patient populations.

Abbreviations

COC

cumulus-oocyte-complex

E₂

estradiol

ECM

extracellular matrix

HA

hyaluronic acid

iPSCs

induced pluripotent stem cells

IVF

in vitro fertilization

IVGM

in vitro follicle growth and maturation

IVM

in vitro maturation

MII, metaphase II; OTC

ovarian tissue cryopreservation

OTT

ovarian tissue transplantation

VEGF

vascular endothelial growth factor

Funding

This work was supported, in part, by the National Institutes of Health, Eunice Kennedy Shriver National Institute of Child Health and Human Development (NIH, NICHD; award F30HD107966 to E.L.T.; NIH, NICHD T32HD094699 to H.B.M.; NIH, NICHD award R01HD104683 to E.L.T., H.B.M., and M.M.L.; NIH, NICHD award R21HD108710 to E.L.T., H.B.M., and M.M.L.), and the Gesualdo Foundation Research Scholar funds (to M.M.L.).

Disclosures

The authors have nothing to disclose.

Data Availability

Data sharing is not applicable to this article as no data sets were generated or analyzed during the current study.