Effects of intraovarian injection of autologous platelet rich plasma on ovarian reserve and IVF outcome parameters in women with primary ovarian insufficiency

Introduction

Ovarian aging is a physiological process associated with a decline in the quantity and quality of the oocytes stored within the follicular cohort [1]. Age-related physiological decline in the number of follicles has critical implications for fertility and is an increasingly more prevalent reason for women to seek fertility treatment. Indeed, the number of women undergoing infertility treatment with assisted reproductive technologies (ART) in the United States with a diagnosis of diminished ovarian reserve (DOR) increased from 22,089 in 2010, to 40,883 in 2017 [2, 3]. The extreme form of DOR, primary ovarian insufficiency (POI), affects 1% of reproductive age women, and is characterized by a severe decrease in ovarian reserve prior to 40 years of age, resulting in menopausal serum gonadotropin hormone levels and menstrual irregularity or amenorrhea [4]. The European Society of Reproductive Medicine and Embryology guidelines define POI as the presence of oligomenorrhea-amenorrhea for at least 4 months and serum follicle stimulating hormone (FSH) levels of ≥ 25 IU/ml measured at least twice with a 4-week interval, with an onset before the age of 40 years [5].

Currently, oocyte donation is the only established treatment modality for women with POI [6]. Similar to the trend observed in women diagnosed with DOR, number of women undergoing oocyte donation in the United States increased from 15,504 in 2010 to 21,033 in 2017 [2, 3]. While utilization of oocyte donation continues to increase, a large number of women express having difficulty adjusting to the idea of not having a child that carries their own genetic material. Another impediment to oocyte donation is that a number of countries have limitations regarding the use of donor oocytes due to ethical or religious concerns, compelling women to seek experimental therapies.

At present, there are no effective treatment modalities for women with POI who wish to have children with their own eggs. While a number of experimental strategies have been tested, live births have been difficult to achieve in women with POI [7]. Kawamura et al. have performed Hippo pathway disruption and Akt activation [8]. First, they have removed ovaries from women with POI and cryopreserved ovarian cortical strips. They then fragmentated the ovarian strips to disrupt Hippo signaling, stimulated Akt signaling in-vitro, and grafted the fragments back to the ovaries. In the first study using this approach involving 27 women, they reported a live birth after in vitro fertilization (IVF) and transfer of an embryo. In a follow up study from the same group, among 37 women with POI, 9 achieved follicle growth in auto-grafts, 6 underwent oocyte retrieval, and 4 underwent embryo transfer, resulting in two live births (5.4%) [9]. Experimental treatment modalities have also been used in women with DOR. Herraiz et al. have investigated the effects of autologous stem cell ovarian transplant (ASCOT) on ovarian reserve in poor responders [10]. In this study, bone marrow derived stem cells (BMDSC) were mobilized to peripheral blood by granulocyte colony-stimulating factor (G-CSF) treatment and subsequently collected by apheresis. Cells were delivered into the ovarian artery by an intra-arterial catheter. Seventeen women were treated, resulting in increased number of antral follicles and oocytes, and five pregnancies, three of which were spontaneous. While ASCOT is a complex procedure, their study is encouraging regarding the potential effect of bone marrow or blood-derived substances on follicle development in women with a limited ovarian reserve.

Platelet Rich Plasma (PRP) consists of a high concentration of platelets found in plasma obtained after centrifugation of peripherally collected blood [11]. Platelets carry more than 800 types of protein molecules, cytokines, hormones, and chemoattractants [12]. The activation of platelets induces the release of a variety of biologically active proteins, which stimulate cell proliferation, growth, and differentiation [13]. Biologically active mediators relased by activated platelets include platelet-derived growth factor (PDGF), transforming growth factor-β (TGF-β), fibroblast growth factor (FGF), epidermal growth factor (EGF), insulin-like growth factor-1 (IGF-1), and hepatocyte growth factor (HGF) [14]. PRP induces accelerated angiogenesis and anabolism, inflammation-control, cell migration, differentiation and proliferation [15]. Autologous platelets have been used as a source of proteins for tissue healing and regeneration and have been suggested to promote the development of isolated human primordial and primary follicles to the preantral stage [16].

In this study, we hypothesized that intraovarian injection of autologous PRP would improve ovarian reserve parameters, ovarian response to stimulation, and IVF outcomes in women diagnosed with POI. We treated 311 women previously diagnosed with POI with intraovarian injection of PRP. Of these women, 23 (7.4%) achieved spontaneous pregnancy, and another 82 (26.3%) developed at least one cleavage stage embryo after controlled ovarian hyperstimulation (COH) and IVF.

Results

A total of 311 women (mean age ± SD: 34.8 ± 4.3) with the diagnosis of POI were included in the study. Flowchart of outcomes is shown in Figure 1.

Discussion

In this study, we investigated whether intraovarian injection of autologous PRP would improve ovarian reserve parameters, ovarian response to stimulation, and result in live births in women diagnosed with POI. Intraovarian PRP injection was performed in a total of 311 women. Among them, 25 women (8.0%) achieved livebirth/sustained implantation (either spontaneously or after IVF), while another 25 (8.0%) stored cryopreserved embryos.

PRP has been utilized and studied since 1970s and has been implemented into routine clinical practice as a rejuvenating agent or to promote healing in dermatology, plastic surgery, dentistry, and orthopedics [17]. Similarly, in obstetrics and gynecology, several studies with small sample size were conducted to investigate the effects of PRP injection into the uterus and ovaries [18]. The first data about intraovarian injection of PRP were published by Sills et al. [19]. They reported improvement in the laboratory parameters after intraovarian PRP in four women with DOR. Subsequently, Sfakianoudis et al. reported the first pregnancy in a menopausal woman after intraovarian PRP injection [20]. This was a 40-year-old woman diagnosed with POI at the age of 35. Her serum biomarkers of ovarian reserve improved after the PRP procedure and pregnancy was achieved after natural IVF; however, miscarriage occurred in the fifth week of gestation. Examining an in vitro model, Hosseini et al. reported a significant increase in the development of early antral follicles after they were cultured with PRP, concluding that it may be an effective method to induce follicular development [16]. Our study builds on these previous reports and using large sample size, provides data that can guide future research and help patient counseling.

In women with the diagnosis of POI, extremely low ovarian reserve parameters are not always associated with absence of follicles in the ovary. Kawamura et al. removed ovarian tissue from 27 women with POI through laparoscopic surgery and cut these tissues into strips [8]. After Hippo signaling disruption and Akt stimulation, small fragmented ovarian strips were autotransplanted beneath the serosa of Fallopian tubes. Follicular growth was observed in 8 of these 27 women (29.6%), all of which developed preovulatory follicles within less than 6 months. Retrieval of mature oocytes occurred in 5 women (18.5% of total), resulting in a livebirth after embryo transfer. In a follow up study from the same group, 37 women with POI were treated similarly, resulting in retrieval of mature oocytes from 6 women (16%), and 2 livebirths (5.4%) [9]. More recently, Kawamura et al. have described a drug-free approach based on Hippo signaling disruption alone by performing partial ovarian cortical removal in women with DOR and autografting back during the same laparoscopic surgery [21]. They treated 11 women with severely diminished ovarian reserve, 9 had an increase in AFC and underwent IVF-ICSI. In addition to one spontaneous pregnancy, embryo transfer resulted in one live birth, and two ongoing pregnancies. Three additional women had embryos cryopreserved for future transfer. In our study, 23 women conceived spontaneously (7.4%), and antral follicle development was observed in 70% of the rest of the study group (201/288). Of these 201 women, 82 generated embryos; 57 underwent embryo transfer and 25 of them cryopreserved embryos to be transferred at a later stage. Therefore, autologous PRP injection seems to result in encouraging outcomes, without the complexity of ovarian surgery.

Spontaneous pregnancy is uncommon in women with POI, with reports of spontaneous conception ranging from 2.2% to 14.2% [22]. It is noteworthy that none of the women in our study group had a history of spontaneous pregnancy before PRP, and the 23 spontaneous pregnancies that occurred after PRP, all happened within the first two cycles after the procedure. Similarly, Pantos et al. recently reported two women with POI aged 27 and 40 years, and one postmenopausal women aged 46 years to conceive spontaneously after PRP injection [23]. Spontaneous pregnancy was also reported by Herraiz et al. in 3 of 17 women with DOR who received autologous bone marrow derived stem cell injection into their ovarian artery [10]. We found that 7.4% of women with POI treated with PRP spontaneously conceived. The mechanisms by which ovarian activation strategies induce spontaneous conception are worthy of further investigation.

PRP contains a number of active substances [11]. However, PRP‘s mechanisms of action in general, and its effects in the ovary in particular, remain largely unknown. Within this context, our study provides relevant information. First, among women who developed at least one antral follicle after PRP injection and were therefore able to undergo IVF in our study, 64% started within the first three cycles and another 21% in the fourth cycle. The progression of human follicles from a primary follicle to ovulation is estimated to require approximately 85 days [24]. This development occurs in gonadotropin independent and dependent phases [25]. Our findings suggest that any type of follicle present in the ovary may be directly affected by PRP. In addition, indirect effects through action on somatic compartment are also possible. Second, we found that women who did not have an antral follicle at the time of PRP injection were less likely to respond to treatment compared to those who had one or two antral follicles (Tables 2, 3). Similarly, women in the lowest quartile for serum AMH and highest for serum FSH, were less likely to respond (Tables 2, 3). Collectively, our findings suggest that PRP helps activate existing preantral and/or early antral follicles, and that the number of remaining follicles in the ovaries of women with POI likely determines the extent of their response. PRP treatment resulted in unchanged levels of serum FSH, with minimal improvement in AMH and AFC (Figure 2), supporting this conclusion.

Among 201 women in whom COH was attempted in this study, 82 (40.8%) had at least one cleavage embryo developed. While this is a very encouraging result, when 55 of these 82 women underwent embryo transfer, only 9/55 (16.3%) achieved sustained implantation and livebirth. This attrition could be due to a number of factors. First, at the center where the study is conducted, embryos of women with POI are often transferred or cryopreserved at the clevage stage, associated with a lower implantation rate compared to blastocysts. Second, preimplantation genetic testing for aneuploidy (PGT-A) was not used, therefore aneuploid embryos were not excluded from transfers. Finally, it is possible that activation of otherwise dormant follicles in women with POI might recruit oocytes with lower viability.

This study has a number of limitations. First, the pregnancy rates after embryo transfer could not be conclusively established as some of the women conceived spontaneously and others opted to store cryopreserved embryos. Second, in this study we compared patients’ outcomes to their pre-treatment state. Therefore, our study does not include an independent control group and is not randomized. The description of a control group is also open for debate. Based on the work by Kawamura et al. [8, 21], one might ask whether the observed effect of PRP could be primarily mechanical. In the same context, Zhang et al. have evaluated follicle development and pregnancy outcome in 80 women with POI after ovarian biopsy/scratch [26]. They reported that they obtained mature oocytes in 10 women (12.5%) and one (1.25%) delivered a healthy singleton baby after IVF. At present, there are no existing data on sham/vehicle injections as controls. Clinical delivery of cells or cell products to the ovary (directly or via the ovarian artery) may be having their effects due to the injection, or, due to vehicle effects, and not biomolecules or cells. Until specific molecules that have reproducible effects are identified, such a mechanism for PRP effect should be considered as a possibility [26].

In conclusion, in women with POI, intraovarian injection of autologous PRP might be considered as an alternative experimental treatment option. Future studies with larger sample size and randomized prospective study design are necessary to determine whether this intervention truly results in improved clinical outcomes, through spontaneous conception and/or assisted reproduction. Until then, autologous PRP treatment should not be recommended as part of routine treatment in women with POI.

Materials and Methods

Supplementary Materials

Supplementary Table 1