Abstract
To review the outcomes of in vitro maturation (IVM) and in vitro fertilization (IVF) in women with empty follicle syndrome (EFS). The study evaluated the genetic underpinnings of EFS by analyzing mutations.
This retrospective case series involving 17 women with EFS over at least 2 IVF cycles was conducted. The study also employed whole-exome sequencing to analyze the genetic mutations. The treatment approaches included letrozole-primed IVM, follicle-stimulating hormone (FSH)-human chorionic gonadotrophin (hCG)-primed IVM, and conventional IVF.
The average female age was 31.5±4.6 years, and the duration of infertility was 7.3±3.5 years. Four patients underwent IVF. IVM oocyte collections yielded oocytes in 12 of 13 subjects. Of these, 75% (9/12) yielded MII oocytes after 48 h of IVM media incubation. Six subjects had fertilized embryos, resulting in a 40.9% intracytoplasmic sperm injection (ICSI) fertilization rate (9 embryos/22 MII oocytes). Genetic analysis revealed mutations in seven patients. This study demonstrated the partial efficacy of letrozole-primed IVM plus growth hormone and FSH-hCG primed IVM protocols. No pregnancies or live births were recorded after IVM. One ongoing pregnancy post-IVF and one spontaneous live birth were observed.
Inter-cycle variabilities were observed in women with oocyte maturation abnormalities (OMAs). Almost all patients with EFS had oocytes collected during IVM following IVF. These oocytes have limited potential for maturation, fertilization, and live birth, as demonstrated by the low rates observed after IVM culture and ICSI. These conditions are observed in OMAs due to defects in the oocyte machinery. The proposed flowchart provides a comprehensive classification approach for various forms of EFS.
PRECIS: Empty follicle syndrome.
Introduction
Failure to yield oocytes from fully developed follicles at oocyte pick up (OPU) 35-37 hours after human chorionic gonadotropin (hCG) injection during an in vitro fertilization (IVF) cycle is defined as empty follicle syndrome (EFS)(1). EFS has been a subject of intense debate regarding whether it is a true entity or failure of IVF cycle management. Before the implementation of gonadotropin-releasing hormone analogs (GnRHa) for triggering in antagonist IVF cycles, EFS was subdivided into two subtypes: The first is known as false-EFS (fEFS), which is due to the lack of hCG application before OPU or low efficacy of applied hCG; the second is referred to as genuine-EFS (gEFS), which is not related to errors in hCG but a repeated failure in at least two consecutive cycles to collect oocytes despite adequate dose and timing of hCG application(2).
The prevalence of gEFS ranges from 0.045% to 7% in the literature(3-6). Hourvitz et al.(7) published the first successful pregnancies and live births via in vitro maturation (IVM) in women with gEFS, the cause of which was accepted as part of the oocyte maturation abnormalities (OMAS). Human genetic studies have revealed that mutations related to oocyte maturation and zona pellucida (ZP) play crucial roles in oocyte development(8). Most cases of EFS were found to have mutations related to ZP proteins(9-11). To date, more than 25 mutations in ZP-related proteins have been reported(12).
For follicular growth and oocyte developmental competence, the corona radiata projections and oocyte microvilli must be in contact with the ZP. Absence of, or deficiency of ZP formation impairs bidirectional communication between the oocyte and surrounding cumulus cells and diminishes oocyte development by destabilizing the cumulus, and affected subjects may develop female infertility(13-15). Defects in ZP, especially those related to integrity rather than thickness, result in a lack of oocyte development, defective folliculogenesis, and early granulosa cell apoptosis, which may explain the pathophysiology of EFS(16). Hatirnaz et al.(17) previously reported intracycle and intercycle variabilities in women with OMAs(17). Among patients who had undergone 2-8 previous IVF attempts, some experienced EFS in at least two cycles and experienced oocyte maturation arrest and biochemical pregnancies in different cycles. This observation suggests that the follicles are not empty in women diagnosed with gEFS, but rather are due to oocyte failure. It is noteworthy that even in women with gEFS associated with genetic mutations, there may be potential for improvement in oocyte yield and maturation. EFS can be redefined as 1) EFS as part of OMAs, 2) Genuine EFS (gEFS) (altered gene expressions, or early apoptosis, etc.), 3) GnRHa use-related EFS, and 4) hCG-related EFS (fEFS). FEFS is a poorly researched condition that affects some women undergoing IVF treatment. HCG-treatment-related errors appear to be the main mechanism underlying fEFS. Improper hCG administration has often been the most common cause(3). Some studies have suggested that increasing the dose of gonadotropins, administering a second (rescue) dose of hCG, or extending the duration from hCG injection to oocyte collection may improve the chances of retrieving oocytes.
The effectiveness of these approaches remains unclear and requires further investigation(18). A potential preventative measure of serum hCG level the day after the trigger and a second bolus of hCG administration has been proposed to prevent against fEFS(5). Clinicians can trigger GnRHa, especially in hyperresponder patients, because of the shorter surge effect of GnRHa compared with urinary or recombinant hCG(19-21). The shorter duration of luteinizing hormone (LH) activity observed in GnRHa cells substantially limits the risk of OHSS but may also contribute to diminished oocyte performance, even leading to false EFS(22). The mechanism of fEFS in this case was different from that observed with hCG, where GnRH agonists induce the release of LH and follicle-stimulating hormone (FSH) from the pituitary. Therefore, with any temporary or permanent factor leading to the dysfunction of the pituitary gland (i.e., hypogonadotropic-hypogonadism), the expected flare action of the analog can be altered and may be responsible for diminished oocyte yield, maturation problems, and EFS(23). For cases with a history of previous EFS, a dual trigger can be planned. During GnRH agonist-triggered cases, especially for patients with the predisposing factors listed subsequently, if no oocyte is yielded from an ovary, the procedure can be stopped, and a second trigger is planned, usually consisting of hCG. Here we performed a case series, identified 17 women with gEFS who had letrozole-primed IVM plus growth hormone (GH), FSH-hCG-primed IVM treatment, and IVF, and investigated oocyte maturation and embryo development, as well as pregnancy outcomes in women who had repetitive EFS after undergoing at least two IVF cycles. Additionally, in this paper, we introduce a flowchart for managing the multiple forms of gEFS. This study aimed to review IVM and IVF treatment outcomes in women with EFS and evaluate the genetic underpinnings of EFS by analyzing mutations.
Materials and Methods
In this case series study, 17 women who had gEFS in their previous (at least two) IVF cycles were detected. Only patients who had gEFS after hCG trigger were included in the study group. FEFS and EFS secondary to GnRH agonist triggering were not included in the study. Demographic characteristics and basal hormonal parameters were all evaluated and recorded. The 17 women were contacted, and written informed consent was obtained from the women studied. This study was approved by the Clinical Research Ethics Committee of Samsun University, Turkey (decision date: 26.04.2023, decision number: 2023/8/20).
Letrozole-priming IVM Plus Growth Hormone
Letrozole 2.5 mg tablets (PO) twice daily started on day 3 of the menstrual cycle for 5 days following the initial transvaginal ultrasound examination. On day 7, a second transvaginal ultrasound examination was performed to measure follicular growth and endometrial thickness. Recombinant GH (Genotropin 36 IU/12 mg) 2 mg/day subcutaneously was added to the treatment regimen on day 10 for 6 days. A third ultrasound examination was conducted to assess the antral follicles. Once at least 2-3 follicles reached a diameter of 10-12 mm and the endometrial thickness reached 8 mm, a 6.500 IU subcutaneous injection of hCG (Ovitrelle, Merck Serono, Turkey) was administered 38 h prior to oocyte collection. Oocyte retrieval was performed under an aspiration pressure of 80 mmHg using an 18-gauge aspiration needle with a double lumen for continuous flushing. Embryos obtained in this protocol were frozen either at the cleavage or blastocyst stages, and a frozen-thawed embryo transfer cycle was performed 1-2 months after the procedure.
FSH-hCG-priming IVM
Recombinant FSH (Gonal-f, Serono) was administered subcutaneously at a daily dose of 75-150 IU, starting on day 3 of the menstrual cycle and continuing for 3 days following the initial transvaginal ultrasound examination. On day 7, a second transvaginal ultrasound examination was performed to measure follicular growth and endometrial thickness. Around day 9-10, a third ultrasound examination was performed to measure the size of the antral follicles. When follicles reach 10-12 mm in diameter and the endometrial thickness reached 8 mm, recombinant hCG 6500 IU (Ovitrelle, Merck Serono) was administered 38 h before oocyte retrieval. Oocyte retrieval was performed with a 17-18-gauge aspiration needle with a double lumen. Embryos obtained in this protocol were transferred to the fresh cleavage stage, and the remaining embryos, if any, were frozen at the cleavage stage.
Luteal Phase Support
The luteal phase was supported with estradiol valerate at a dose of 6-8 mg per day until fetal heartbeat was detected during ultrasound examination. Progesterone 200 mg capsules (800 mg/day (Koçak Farma, İstanbul, Turkey) intravaginally were used routinely in both letrozole primed IVM plus GH and in FSH-hCG IVM cycles until the 12th week of gestation.
Laboratory Procedures
Laboratory procedures for IVM were performed using both methods according to a modified protocol, as previously described(24). The authors have been using MediCult IVM culture systems (CooperSurgical, USA) since 2007. A SAGE Vitrification Kit (Cooper Surgical, USA) was used for the vitrification and warming procedures.
The maturation process was assessed at 26 and 48 h. Oocytes that had matured at 26 h underwent intracytoplasmic sperm injection (ICSI) as further culture was deemed unnecessary. Any remaining immature oocytes were maintained in IVM culture until 48 h. Oocytes that matured at 48 h were then subjected to ICSI. The cleavage-stage embryos were graded according to the system reported by Hsu et al.(25). Blastocysts were graded according to a previously reported system by Neuber et al.(26).
A pregnancy test was administered on either day 10 post-blastocyst transfer or day 12 following cleavage-stage embryo transfer. Serum hCG levels exceeding 5 IU/L during the initial assessment indicated a positive pregnancy. Clinical pregnancy was defined as the detection of an intrauterine fetus with a heartbeat on ultrasound. Ongoing pregnancy was defined as the pregnancies over 12th weeks of gestation. A spontaneous abortion is defined as the presence of an empty gestational sac or a gestational sac containing an embryo or fetus without fetal heart activity within the first 12 weeks of gestation. Live birth was defined as the birth of a living child born after 28 gestational weeks. Falling beta hCG levels without visualization of pregnancy on ultrasound were considered as biochemical pregnancy loss.
Whole-exome Sequencing
Genetic mutations in 9 out of 17 patients with gEFS were analyzed by whole-exome sequencing (WES). Briefly, DNA was extracted from whole blood samples using a QIAmp kit (Hilden, Germany) in a QIAcube HT system (Hilden, Germany) following the manufacturer’s instructions (Richards et al. 2015). Library preparation/comparison was conducted using a QIAseq Index I set A kit (Hilden, Germany) in combination with a custom-designed targeted NGS panel (CDHS-15607Z-1008) for female infertility-associated genes.
Statistical Analysis
Statistical analyses were performed using SPSS software (IBM SPSS Statistics, Version 21.0, USA). Descriptive statistics, including means and standard deviations, were calculated to summarize the demographic and baseline clinical characteristics of the study participants.
Results
The average age of the 17 female participants was 31.5±4.6 years (range: 24-38), and the average duration of infertility averaged 7.3±3.5 years (range: 1.5-15.0). Basal serum measurements revealed FSH levels at 8.24±3.56 IU/L, LH levels at 8.93±6.40 IU/L, estradiol levels at 59.67±44.48 pg/mL, progesterone levels at 0.58±0.29 ng/mL, thyroid stimulating hormone levels at 1.93±0.99 mIU/mL, T3 levels at 2.96±0.40 pg/mL, T4 levels at 1.16±0.25 ng/mL, anti-Mullerian hormone levels at 2.96±2.03 ng/mL, PRL levels at 21.17±9.74 ng/mL, and antral follicle count of 17.4±8.9. Notably, four (out of the 17) patients did not undergo the IVM procedure. IVM procedures resulted in successful oocyte collection in 12 of 13 subjects. Among these, 9 out of 12 subjects (75%) yielded MII oocytes after 48 h of incubation with IVM medium (with a range of mature oocytes in those who matured being 2-8). Only 6 subjects ended up with fertilized embryos (9 of 22 MII oocytes total. This represents a fertilization rate of 40.9% with ICSI. None of the subjects in this study had experienced any IVM pregnancies. One patient achieved an ongoing pregnancy following an IVF cycle (Case 5), while another case (Case 2) resulted in a spontaneous live birth.
The detailed demographic, clinical, and laboratory characteristics of each patient are presented in Table 1.
The application of IVF and IVM treatments and outcomes are outlined in Table 2.
Wes analysis was performed on 9 subjects. Genetic mutations (FSHR mutations, TACR3 mutation, LHX4, mutation, STAG3 mutation, ZP1 mutation, ZP3 mutation, PATL2 mutation and LHCGR mutation) were detected in 7 women. Cases 7 and 8 had no mutation in WES analysis. In eight patients (Case 1, 3 9, 10, 14-17), WES analysis was not performed, as indicated in Table 3.
- In Case 2, a heterozygous mutation in STAG3 was identified on Exon 4, specifically at nucleotide position c.319G>A. This mutation follows an autosomal recessive (AR) inheritance pattern.
- Case 4 exhibited a homozygous mutation in the LHCGR gene located in Exon 11 at nucleotide position c.970_971del A>G, consistent with AR inheritance.
- Case 5 was found to have a novel homozygous mutation in the ZP1 gene in Intron 11, denoted as c.1775.3C>G, which also followed an AR inheritance pattern.
- A more complex genetic profile was observed in Case 6, which exhibited three mutations: a homozygous mutation in the FSHR gene on Exon 10 (p. S680N), a heterozygous mutation in the same gene but at a different site (Exon 10, p. A307T), and additional heterozygous mutations in the TACR3 gene (Exon 2, c.737c>t) and ZP3 gene (Exon 2, c.382A>C). The FSHR and TACR3 mutations follow AR inheritance, whereas the ZP3 mutation is autosomal dominant.
- Case 11 showed a heterozygous mutation in LHX4 located at c.1158T>A, which followed an AR inheritance pattern.
- In Case 12, two homozygous mutations were detected in PATL2: one in Exon 7 (c.320C>T) and the other in Intron 7 (c.446+1G>C), both consistent with AR inheritance.
- Case 13 was identified with a homozygous mutation in ZP1 on Exon 12, which was noted as C1775-3C>A, following an AR pattern.
Detailed data are presented in Table 3.
Discussion
Our paper introduces several novel insights into the study of EFS. First, we present EFS as a subtype of OMAs that is determined by genetic mutations through a comparative study. Second, it explores the role of a letrozole-primed plus GH IVM protocol in women with gEFS. OMAs often result in the collection of 100% immature oocytes, even after triggering with hCG, GnRH agonist, or both(27, 28). IVM is performed to collect oocytes from follicles measuring 2-10 mm, smaller than the size in IVF cycles. IVM, being independent of the development of LH receptors in the follicle, as required in IVF, could serve as a treatment for EFS.
The study investigated 17 women diagnosed with gEFS. Out of the 9 women analyzed by WES, seven had various mutations noted (FSHR mutations, TACR3 mutation, LHX4, mutation, STAG3 mutation, ZP1 mutation, ZP3 mutation, PATL2 mutation and LHCGR mutation). Two women did not present with mutations in their WES analysis. The results of this study indicated that IVM for gEFS, whether using the letrozole-primed IVM + GH protocol or FSH-hCG primed IVM, may result in oocytes and embryos in rare cases but is not likely to result in a live birth. This study also demonstrated for the first time that EFS is a subtype of OMA and that genetic testing for mutations is highly recommended if available.
Since its first report, EFS has been a hot topic of debate in the field of assisted reproductive technology(1). EFS is classified as g EFS, in which repeated failure to retrieve oocytes with appropriate blood hCG levels is observed, and fEFS, in which a low blood level of hCG (<40 IU/L) is detected either due to misuse of hCG for triggering or low bioavailability of the HCG used(29, 30). Filtrates of follicular fluids collected from patients diagnosed with gEFS during a stimulated IVF cycle were investigated, and it was demonstrated that immature oocytes were present in the filtrates and that these could mature in vitro using IVM culture media(31). This finding, though it is only a case report, showed that the follicles were not empty. Inan et al.(32) reported the whole gene expression profile of granulosa cells from the follicular fluid of a woman who underwent three repeated gEFS and showed that around 160 genes were differentially expressed. This study is the first to demonstrate the role of gene expressions that trigger early apoptosis in healthy oocytes and fail maturation. Thin ZP was associated with early apoptosis of oocytes present in 200 preantral follicles. Ovarian aging or low ovarian reserve is also a cause of gEFS(32, 33). A possible genetic cause of gEFS was reported by Onalan et al.(34), in which two sisters with moderate sensorineural deafness were diagnosed with gEFS due to alterations in the transient and sequential expression of epidermal growth factor, which is essential in the meiotic resumption process. The recurrence rate of EFS was reported to be 15.8%(35). Baum et al.(35) found that patients with recurrent EFS exhibited significantly prolonged infertility and lower estrogen levels compared with those with sporadic EFS. However, their study did not include hCG hormone analysis; thus, the differentiation of EFS subtypes was not performed. Additionally, no genetic screening was performed for either sporadic or recurrent cases. Therefore, elucidating the underlying cause is challenging based on these data.
Mutation studies related to female infertility have revealed the involvement of ZP1-4, LHCGR, STAG3, and mutations related to premature ovarian failure (POF) that play a role in the pathogenesis of EFS. Thus, EFS, POF, and resistant ovary syndrome were included as OMA subtypes.
The significance of EFS in relation to mutations in the LHCGR gene(36, 37). We identified an LHCGR mutation in one of our EFS cases (Case 4) in which IVM resulted in oocyte retrieval and embryo transfer, but pregnancy was not achieved.
Furthermore, the relevance of EFS has been emphasized in connection with mutations in the ZP1 gene(38, 39). In our study, a noteworthy observation was made regarding the ZP1 mutation within a consanguineous family, specifically involving three sisters. It exhibited distinct manifestations in the three sisters, particularly in the eldest sister (Case 12), which was included in our study. These manifestations included EFS, oocyte maturation arrest, and the production of mature oocytes with successful embryo transfer in one cycle. Additionally, her IVF cycles produced zona-free oocytes. The genetic analysis of this case was performed at the outer center. Although we have a report in our archives indicating the presence of a ZP1 mutation, we do not have the details. Therefore, it could not be included in Table 3. The genetic profile of the family has been published(40). Additionally, in our WES analysis, we identified two distinct PATL2 mutations in this case. We also observed patients with gEFS during IVF cycles while having spontaneous biochemical pregnancies and sometimes mature oocytes during other IVF cycles. This finding is novel, and it is difficult to determine whether these patients have EFS or not, but we can accept these cases as subtypes of OMAs.
According to our WES analysis, two distinct mutations in ZP1 were identified in cases 5 and 13. The mutation detected in Case 5 is novel; however, it is located within Intron 11 of the ZP1 gene, suggesting no phenotypic impact. Case 5 was diagnosed with EFS as part of OMAs. In the initial two IVF attempts, no oocytes were obtained. In the third IVF attempt, only one zona-free oocyte was retrieved, and ICSI was performed, resulting in embryo freezing. In the fourth IVF attempt, two normal oocytes were retrieved. Successful ongoing pregnancy was achieved through embryo transfer. Case 13, who was also diagnosed with EFS as part of OMAs, experienced this condition in two IVF attempts. In another IVF attempt, a germinal vesicle arrest. Both cases did not receive IVM treatment.
Moreover, the emphasis on EFS has been placed in the context of mutations in the ZP3 gene(41, 42). Within our study, Case 6 was diagnosed with premature ovarian ınsufficiency (POI) and true EFS. Additionally, zona-free oocytes were observed in some cycles. Intriguingly, based on WES analysis, one mutation in ZP3 and two mutations in FSHR. The patient underwent four IVF attempts, with oocytes failing to be retrieved in three of the cases. In the fourth attempt, an oocyte was obtained, but it degenerated. Two attempts of IVM resulted in the retrieval of one oocyte each time; however, subsequent observations revealed zona-free oocytes and subsequent degeneration in the germinal vesicle stage.
In one of our cases (Case 2) with g-EFS in her previous two consecutive IVF cycles, we determined a STAG3 mutation in her WES analysis (this mutation was previously a variant of uncertain significance but likely benign according to Association for Clinical Genomic Sciences classification). She had a history of biochemical pregnancy prior to IVF attempts and was diagnosed with POI. We performed IVM cycles, obtained embryos, and transferred them twice but failed to achieve pregnancy. However, the patient had three consecutive spontaneous biochemical pregnancies while waiting for another frozen-thawed embryo transfer. The last result resulted in the delivery of a healthy baby. The clinical course of the patient was interesting, beginning with POI and EFS and ending with spontaneous pregnancy.
When we looked at the cases studied, apart from the FSH-hCG primed IVM group, in which all cases were gEFS, the remaining nine women treated with letrozole-primed IVM + GH had intercycle variabilities either in their previous IVF cycles or in their IVM cycles. This finding is important and explains why EFS has follicles that are not empty, providing new evidence that EFS is a subtype of OMAs(17, 27).
We propose a novel flowchart for the clinical management of all types of EFS (Figure 1). Apart from gEFS, Functional EFS, and GnRH-related EFS, we describe a novel subtype: EFS as a subtype of OMAs. In managing this entity, repeating the cycle for accurate diagnosis or directly investigating mutations via WES may be recommended. Because of the importance of intercycle variability, mutation analysis before a second attempt is preferable.
Study Limitations
Our study has some limitations. The small sample size and retrospective design may limit the generalizability of our findings. Therefore, studies with larger series should be scheduled to draw proper conclusions regarding the use of letrozole IVM plus GH therapy.
Conclusion
Almost all patients diagnosed with EFS after IVF had oocytes collected during IVM, indicating that the follicles were not truly empty. These oocytes have limited potential for maturation and fertilization, as evidenced by the low success rates following IVM culture and ICSI. The conditions observed in these OMAs stem from defects in the oocyte machinery, resulting in oocyte and cytoplasmic immaturity. Despite these challenges, pregnancies did occur in some subjects, suggesting that EFS can occasionally be overcome. The current evidence does not support the success of IVM in patients with g-EFS; however, the number of cases studied is limited, and further research is needed.
