Spontaneous ovarian hyperstimulation in a nonpregnant woman with PCOS: a rare case highlighting FMN2 missense mutation and androgen receptor gene deletion

Abstract

Background

Polycystic ovary syndrome (PCOS) is a common endocrine disorder characterized by oligo-anovulation, hyperandrogenism, and polycystic ovarian morphology. While its etiology is multifactorial, genetic susceptibility plays a key role in its pathophysiology.

Case presentation

We report a rare case of spontaneous ovarian hyperstimulation syndrome (sOHSS) in a 26-year-old nulligravid Japanese woman with PCOS who had never received exogenous gonadotropins. She presented with amenorrhea, elevated luteinizing hormone (LH), and bilaterally enlarged ovaries mimicking OHSS. Laparoscopic ovarian drilling (LOD) was performed, resulting in marked LH reduction, restoration of ovulatory cycles, and spontaneous pregnancy. Genetic analysis revealed a novel heterozygous missense mutation in the FMN2 gene (c.3461 C > T, p.Pro1154Leu) and a heterozygous deletion in exon 1 of the androgen receptor (AR) gene, detected in peripheral blood, bilateral ovarian tissue, and granulosa cells. No mutations commonly associated with PCOS or OHSS were found. During pregnancy, bilateral ovarian enlargement progressed, and at 33 weeks, magnetic resonance imaging revealed ovarian herniation into the pouch of Douglas. Due to concern relating to a possible mechanical dystocia, an elective cesarean delivery was performed at 38 weeks, resulting in the birth of a healthy infant.

Conclusion

This is the first report identifying FMN2 and AR gene variants in a patient with sOHSS and PCOS, suggesting that noncanonical rare mutations may underlie atypical phenotypes and contribute to ovarian dysfunction in the absence of exogenous stimulation.

Background

Polycystic ovary syndrome (PCOS) is a common endocrine disorder affecting 6–13% of reproductive-aged women, depending on diagnostic criteria [1]. It is characterized by oligo-anovulation, hyperandrogenism, and polycystic ovarian morphology. While its etiology is multifactorial, genetic susceptibility plays a key role in its pathophysiology. Genome-wide association studies (GWAS) and candidate gene analyses have identified several risk loci across major biological pathways, including: follicle stimulation receptor (FSHR) and luteinizing hormone/choriogonadotropin receptor (LHCGR), involved in gonadotropin signaling; cytochrome P450 family 17 subfamily A member 1 (CYP17A1), and DENN domain containing 1 A (DENND1A), which mediates steroidogenesis; insulin receptor (INSR) and thyroid adenoma associated (THADA) with insulin signaling; and Anti-Müllerian Hormone (AMH), implicated in follicular regulation [2–4]. These findings support the notion that PCOS arises from complex interactions among endocrine, metabolic, and follicular, regulatory mechanisms.

Ovarian hyperstimulation syndrome (OHSS) is a systemic condition characterized by massive ovarian enlargement and increased vascular permeability, resulting in fluid shifts from intravascular to extravascular compartments. Vascular endothelial growth factor (VEGF) signaling is a central mediator of this increased vascular permeability, and aberrant VEGF expression has been linked to OHSS severity [5, 6]. Two distinct forms exist: iatrogenic OHSS, occurring in 0.3–5% of patients undergoing in vitro fertilization following human chorionic gonadotropin (hCG)-triggered ovulation [7], and spontaneous OHSS (sOHSS), which develops during pregnancy regardless of the conception method. While iatrogenic OHSS etiology is well-established, sOHSS remains poorly understood. sOHSS, also referred to as hyperreactio luteinalis, is typically associated with excessive hCG production or missense mutations in the FSHR gene that confers increased sensitivity to hCG ². Notably, sOHSS in nonpregnant women has not been reported, suggesting alternative mechanisms beyond FSHR mutations when hCG-induced triggers are absent. Emerging data also implicate TRPC1-dependent calcium signaling—potentially attenuated by nifedipine—and inflammatory chemokine axes (CX3CL1/CX3CR1 and CCL4) in driving ovarian hyperpermeability and stromal remodeling during OHSS, expanding the mechanistic framework beyond classical gonadotropin–VEGF pathways [8, 9].

We present a rare case of PCOS complicated by sOHSS in a nonpregnant state that was successfully managed with laparoscopic ovarian drilling (LOD). Genomic analysis identified variants in androgen receptor (AR) gene and formin homology protein family gene (FMN2) variants. The patient conceived spontaneously and delivered a healthy infant at term.

Case presentation

A 26-year-old nulligravid Japanese woman presented with secondary amenorrhea lasting over three years and a desire to conceive. Menarche occurred at 13 years, followed by irregular menses and progressive oligomenorrhea. She had no family history of endocrine or reproductive disorders. Her height was 159 cm; weight, 47.7 kg; and body mass index (BMI), 18.9 kg/m². Physical examination revealed moderate hirsutism (Ferriman–Gallwey score: 8), with no signs of acne or galactorrhea. Pelvic examination showed a normal-sized, anteverted uterus with bilateral adnexal masses. Transvaginal ultrasound (Tv-US) demonstrated markedly enlarged ovaries (~ 6 cm in diameter bilaterally) with > 12 small peripheral follicles measuring 2–9 mm, consistent with polycystic ovarian morphology.

Diagnostic assessment

Laboratory tests revealed: FSH 8.6 mIU/mL, LH 21.7 mIU/mL, LH/FSH ratio: 2.52, thyroid-stimulating hormone 4.06 µIU/mL, triiodothyronine 3.3 pg/mL, thyroxine 1.1 ng/mL, prolactin 22.0 ng/mL, total testosterone 21 ng/dL, hCG < 0.5 mIU/mL, AMH 6.44 ng/mL, fasting insulin: 3.1 µU/mL, fasting glucose: 75.8 mg/dL, and homeostasis model assessment of insulin resistance: 0.58. She fulfilled all three Rotterdam criteria for the diagnosis of PCOS: oligo/anovulation, clinical hyperandrogenism, and polycystic ovarian morphology [10]. However, her bilateral ovarian enlargement and elevated LH in the absence of hCG exposure or pregnancy raised suspicion of sOHSS. Pelvic T2-weighted magnetic resonance imaging (MRI) (Fig. 1A and B) showed bilateral ovarian enlargement (~ 8 cm), with multiple cysts but no solid masses, ascites, or abnormal enhancement. Brain MRI showed no evidence of pituitary adenoma or hypothalamic abnormalities (Fig. 1C).

Fig. 1.

Images of pelvic and brain MRI and laparoscopic ovarian drilling.A Coronal T2-weighted pelvic MRI showing bilateral ovarian enlargement consistent with OHSS-like morphology. B Axial T2-weighted pelvic MRI showing multilocular cystic enlargement of both ovaries. C Coronal T2-weighted brain MRI showing no evidence of pituitary or hypothalamic lesions. D Laparoscopic view showing bilaterally enlarged ovaries without torsion or hemorrhage. E Monopolar ovarian drilling procedure being performed on the right ovary. F Post-drilling appearance of the ovary following monopolar cauterization and biopsy. MRI: magnetic resonance imaging; OHSS: ovarian hyperstimulation syndrome

Treatment

Given the marked bilateral ovarian enlargement at baseline and elevated LH in the absence of exogenous gonadotropins, pharmacologic ovulation induction was judged to carry a substantial risk of precipitating or exacerbating OHSS. We therefore elected laparoscopic ovarian drilling (LOD) as a non-stimulatory intervention to reduce intraovarian androgen output, lower LH, and restore the FSH/LH balance; an ovarian biopsy was obtained concurrently. Intraoperatively, both ovaries were enlarged (~ 8 cm each), with smooth surfaces and no evidence of torsion, ascites, or pelvic adhesions (Fig. 1D). LOD was performed using monopolar electrocautery (cutting: 40 W; coagulation: 30 W), puncturing each ovary 8–10 times (Fig. 1E and F). Postoperatively, serum LH decreased from 21.7 mIU/mL to 6.5 mIU/mL (a 70% reduction), and spontaneous ovulatory cycles resumed.

Outcome and follow-up

The patient conceived spontaneously two months after LOD. At 7 weeks and 5 days of gestation, Tv-US confirmed a viable intrauterine pregnancy with a crown–rump length of 12.4 mm and a positive fetal heartbeat (Fig. 2A). There was no OHSS flare-up during early pregnancy. However, progressive bilateral ovarian enlargement was noted during pregnancy. At 33 weeks, T2-weighted pelvic MRI revealed that both ovaries had increased to 11–14 cm in diameter, extending into the pouch of Douglas and compressing adjacent pelvic structures (Fig. 2B and C, and 2D). From 33 to 38 weeks, we monitored the patient with serial clinical evaluations (vitals, weight, urine output, edema), ultrasound for ascites (none detected) and ovarian size, and routine fetal surveillance (NST/ultrasound), without maternal complications. Given the mechanical obstruction and risk of dystocia, an elective cesarean section was scheduled. At 38 weeks and 1 day, a healthy male infant was delivered (weight: 2,722 g; length: 50 cm; Apgar scores: 8 and 9 at 1 and 5 min; umbilical artery pH: 7.349). Intraoperatively, bilateral ovarian enlargement was again confirmed without torsion or hemorrhage (Fig. 2E). Postpartum ultrasound revealed persistent ovarian enlargement bilaterally (Fig. 2F). Both mother and infant recovered without complications and were discharged on postoperative day 8.

Fig. 2.

Imaging studies in this case throughout the duration of the pregnancy.A Transvaginal ultrasound at 7 weeks and 5 days of gestation showing a crown–rump length of 12.4 mm with positive fetal heartbeat. B Sagittal T2-weighted pelvic MRI at 33 weeks of gestation. Bilateral ovaries are markedly enlarged with multiple cysts (black arrow: right ovary, white arrow: left ovary), showing impaction into the pouch of Douglas (yellow star: fetal pelvic bone). C Coronal T2-weighted pelvic MRI showing similarly enlarged bilateral ovaries (black arrow: right ovary, white arrow: left ovary) impacted in the pouch of Douglas (yellow star: fetal pelvic bone). D Axial T2-weighted pelvic MRI revealing bilaterally enlarged, multicystic ovaries without any solid components or ascites. E Intraoperative view at elective cesarean delivery at 38 weeks and 1 day of gestation. Both ovaries are markedly enlarged (black arrow: right ovary, white arrow: left ovary) without torsion or hemorrhage. F Postpartum transvaginal ultrasound showing persistent ovarian enlargement. Right panel: right ovary; left panel: left ovary. MRI: magnetic resonance imaging

To investigate underlying genetic factors contributing to the unusual phenotype of PCOS complicated by sOHSS, comprehensive genomic analysis was performed. After obtaining informed consent, DNA was extracted from peripheral blood, ovarian tissue, and granulosa cells obtained during the LOD. Whole-genome sequencing was conducted using a NovaSeq 6000 platform and NEBNext Ultra II DNA PCR-Free Library Prep Kit as described previously [11, 12]. Bioinformatic analysis via CLC Genomics Workbench (Qiagen v24.0.1) identified 10,395 shared variants across all tissues (Fig. 3A). Of these, 209 variants were associated with known PCOS-related genes (Fig. 3B). Comprehensive genomic analysis of peripheral blood, ovarian tissue, and granulosa cells revealed two consistent abnormalities: [1] a heterozygous point mutation in the FMN2 gene (c.3461 C >T, p.Pro1154Leu), and [2] a heterozygous 15-bp deletion (c.1406–1420del) in exon 1 of the AR gene. These variants were detected by mapping the data from whole-genome sequencing (Fig. 4A and B), and both were subsequently confirmed by Sanger sequencing (Fig. 5A and B). These findings suggest a possible genetic basis for the patient’s combined presentation of PCOS and sOHSS. The study was approved by the Institutional Review Board of Kawasaki Medical University (Approval No. 6501-00), and written informed consent was obtained for publication.

Fig. 3.

Comparative Venn diagrams of genetic variants detected in granulosa cells, bilateral ovaries, and blood. A Venn diagram showing the distribution of all detected variants among granulosa cells, left ovary, right ovary, and blood samples. A total of 10,395 variants were shared among all four tissues. B Venn diagram showing the distribution of pathogenic or likely pathogenic variants (classified by Qiagen Clinical Insight) across the same tissues. A total of 209 variants were shared among all samples. Unique and overlapping variant counts are indicated within each region

Fig. 4.

Amino acid sequence alignment showing FMN2 and AR variants across different tissue samples.A FMN2 missense variant p.P1154L detected in all samples. The substitution of proline (P) to leucine (L) at position 1154 is indicated in red. B AR in-frame deletion p.G469_G473del identified in blood, bilateral ovaries, and granulosa cells. The deleted region (five glycine residues) is shown with dashed red lines. This frameshift deletion is likely to significantly impair protein function by disrupting conserved amino acid sequences. AR: androgen receptor; FMN-2: formin 2

Fig. 5.

Detection of FMN2 missense mutation and AR exon 1 frameshift variant by Sanger sequencing A Representative electropherogram showing a reproducible heterozygous missense mutation (NM_020866.4:c.3461 C > T, p.Pro1154Leu) in the FMN2 gene. The variant was confirmed in genomic DNA derived from peripheral blood, ovarian cortex, and granulosa cells. The arrow indicates the C > T nucleotide substitution. B Representative electropherogram showing a heterozygous deletion across exon 1 of the AR gene (AR: Ex1_1406_1420del) resulting in a frameshift from the deleted region. The deletion was detected in DNA from peripheral blood, ovarian cortex, and granulosa cells. The horizontal arrow indicates the deleted segment, and vertical arrows mark the breakpoint boundaries. Absence of expected peaks and the disrupted reading frame suggest a loss-of-function effect. FMN-2: formin 2; AR: androgen receptor

Discussion and conclusions

sOHSS is a rare condition that occurs in the absence of exogenous ovulation induction or pregnancy-related hCG exposure. In nonpregnant individuals, its occurrence is exceedingly rare. This case is particularly notable for persistent bilateral ovarian enlargement and a PCOS phenotype without any external stimulation, prompting suspicion of underlying genetic contributions.

There are currently no internationally established diagnostic criteria for sOHSS. However, this case exhibited several clinical features frequently cited in case reports and review articles as key indicators [13–16]: [1] marked bilateral ovarian enlargement, and [2] polycystic ovarian morphology without any history of ovulation-inducing agents. Based on these findings and the exclusion of neoplastic disease, a clinical diagnosis of sOHSS was considered appropriate.

Numerous genetic variants have been identified in patients with PCOS, whereas reports on sOHSS most frequently implicate mutations in the FSHR gene. In the present case, however, no activating mutation in FSHR—the most reported cause of sOHSS—was detected [13, 17, 18]. Additionally, none of the well-characterized susceptibility loci associated with PCOS identified through GWAS, including FSHR, LHCGR, CYP17A1, INSR, AMH, DENND1A, and THADA, were present [2–4]. Despite the patient exhibiting a PCOS phenotype, the absence of these common variants suggests that rare, noncanonical genetic alterations, such as mutations in FMN2 and AR, may play a pivotal role in the pathogenesis of sOHSS in nonpregnant women.

In this case, genomic analysis of peripheral blood, ovarian tissue, and granulosa cells identified a rare point mutation in the FMN2 gene (p.Pro1154Leu), along with a homozygous deletion in exon 1 of the AR gene. FMN2 encodes formin-2, a cytoskeletal regulator essential for oocyte spindle formation, chromosome segregation, and cellular polarity. While FMN2 dysfunction has been associated with impaired oocyte maturation and infertility in animal models [19–21], its involvement in human reproductive disorders remains largely unexplored. To our knowledge, this is the first report identifying an FMN2 variant in a patient exhibiting both PCOS features and sOHSS. The mutation, located within a conserved functional domain, may disrupt oocyte–granulosa cell communication and follicular architecture, potentially contributing to ovulatory resistance and excessive follicular development. These findings suggest that rare, noncanonical genetic alterations may underlie atypical ovarian hyperresponsiveness, expanding the current understanding of sOHSS pathogenesis beyond the classical FSHR-mediated mechanisms.

In contrast, the AR gene encodes the androgen receptor, a nuclear transcription factor that mediates the biological effects of androgens. Previous studies have demonstrated that AR gene polymorphisms and altered receptor activity are closely associated with hyperandrogenism and the pathophysiology of PCOS, particularly in relation to ovarian follicular development and ovulatory dysfunction [22, 23]. As AR is involved in androgen signaling, this rare variant may account for the PCOS phenotype and hyperandrogenism seen in this case, regardless of traditional susceptibility loci.

LOD is recognized as an effective intervention for women with PCOS, particularly in cases resistant to pharmacologic ovarian stimulation. By modulating the hypothalamic–pituitary–ovarian axis, LOD reduces ovarian androgen production and restores hormonal balance [24, 25]. In the context of sOHSS, where pharmacologic approaches may exacerbate symptoms, LOD offers a nonstimulatory alternative with both therapeutic and diagnostic value. Its application in the present case highlights the potential utility of LOD in managing atypical ovarian hyperresponsiveness, especially when underlying genetic factors are suspected.

This is a rare presentation of sOHSS in the context of PCOS, with pregnancy achieved following ovulation induced by LOD. Genetic analysis revealed no mutations commonly associated with PCOS or sOHSS. Importantly, this is the first report to identify a missense mutation in the FMN2 gene and a deletion in exon 1 of the AR gene in this clinical setting. This case suggests that previously unreported mutations in FMN2 and AR may contribute to the development of PCOS and sOHSS through noncanonical genetic mechanisms. These findings expand current understanding of ovarian dysfunction and highlight the need to consider rare variants in atypical presentations.

Abbreviations

PCOS

Polycystic ovarian syndrome

sOHSS

Spontaneous ovarian hyperstimulation syndrome;

LOD

Laparoscopic ovarian drilling

AR

Androgen receptor

FMN-2

Formin 2

GWAS

Genome-wide association studies

FSHR

Follicle stimulation receptor

LHCGR

Luteinizing hormone/choriogonadotropin receptor

CYP17A1

Cytochrome P450 family 17 subfamily A member 1

DENND1A

DENN domain containing 1A

INSR

Insulin receptor

THADA

Thyroid adenoma associated

AMH

Anti-Müllerian Hormone

Authors’ contributions

K.O. was responsible for surgical management, manuscript preparation, study conceptualization, and genomic data analysis. T.T. contributed to manuscript writing, study conceptualization, and data interpretation. E.N. managed the perinatal care and clinical follow-up. K.W. and K.I. performed the genomic analyses. Y.M. (Yumiko Morimoto) and K.T. assisted with the surgeries. Y.M. (Yukiyo Matsuyama) and T.M. contributed to clinical data acquisition. Y.O. and K.K. supervised manuscript development and provided critical academic guidance. Y.M. (Yoichi Mizukami) conducted genomic analysis, data interpretation, and figure preparation. K.S. supervised the entire study. All authors reviewed and approved the final version of the manuscript.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Data availability

The data that support the findings of this study are available from the corresponding author, T.T., upon reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Institutional Review Board of Kawasaki Medical School (Approval No. 6501-00, approved on August 20, 2024). Written informed consent was obtained from the patient for participation in this study and for the collection and analysis of clinical, imaging, surgical, and genetic data.

Consent for publication

Written informed consent was obtained from the patient for publication of the clinical details and images in this case report.

Competing interests

The authors declare that they have no competing interests.