A dynamic “hormonal tug–of–war” driven by a functioning ovarian remnant after testis-preserving partial gonadectomy in ovotesticular disorders of sex development: a case report

Abstract.

The endocrine behavior of residual ovarian tissue following testis‑preserving surgery in ovotesticular disorders of sex development (DSD) during adolescence remains poorly characterized. We report an adolescent male who presented with Tanner stage B4 gynecomastia despite male‑typical external genitalia. Baseline evaluation revealed elevated gonadotropins, relative hyperestrogenism (estradiol 58.7 pg/mL), and SRY‑negative chromosome mosaicism (47,XX,+der(8)/46,XX). Pelvic imaging demonstrated a small prostate and bilateral intratesticular cysts, and laparoscopy and cystoscopy confirmed a male internal phenotype without Müllerian structures. Because firm intratesticular foci suggested the presence of ovarian tissue, testis‑preserving partial gonadectomy was performed. Histopathological examination revealed bilateral ovotestes, with ovarian follicles adjacent to seminiferous tubules and no OCT3/4‑positive germ cells. Postoperatively, estradiol levels initially decreased but later showed intermittent fluctuations in parallel with testosterone, suggesting hormonal activity from a small residual ovarian remnant. Gynecomastia gradually regressed, bone mineral density remained normal, and serial testicular MRI findings were stable. This case demonstrates that even minimal residual ovarian tissue can remain hormonally active after testis‑preserving surgery in ovotesticular DSD and influence pubertal hormonal dynamics, highlighting the importance of careful long‑term postoperative surveillance with hormonal evaluation and targeted imaging.

Highlights

● Minute ovarian remnants can cause intermittent estradiol-driven oscillations.

● Estradiol surges produce LH/FSH–T fluctuations with imaging and gynecomastia changes.

● Testis-preserving excision maintains endogenous androgen production, while residual ovarian tissue can be monitored using estradiol levels.

Introduction

Ovotesticular disorder of sex development (DSD) is a rare condition characterized by the coexistence of ovarian and testicular tissue within the same individual, typically presenting with a wide spectrum of phenotypic findings. Although most affected individuals present during infancy with ambiguous genitalia, a subset is diagnosed later in childhood or adolescence, including phenotypic males who may remain asymptomatic until puberty (1). In individuals with a fully male external phenotype, ovotesticular DSD may remain unrecognized until adolescence, when it may first manifest as progressive gynecomastia or, less commonly, acute scrotal pain caused by rupture or hemorrhage of an ovarian component within the ovotestis (2). The clinical phenotype is influenced by the relative proportion and functional activity of the ovarian and testicular components, and endocrine manifestations may emerge only when gonadal function increases during puberty.

Management of ovotesticular DSD requires careful consideration of endocrine function, gonadal anatomy, fertility potential, psychosocial context, and long‑term malignancy risk. Surgical strategies generally aim to preserve gonadal tissue concordant with the sex of rearing while removing tissue discordant with the chosen sex (3, 4), particularly when the latter poses endocrine or oncologic risks. In male‑reared individuals with scrotal ovotestes, selective excision of ovarian components can maintain endogenous androgen production and avoid lifelong hormone replacement; however, intraoperative identification of ovarian tissue is challenging when the components are admixed or compartmentalized rather than clearly demarcated (5).

Pubertal activation of the hypothalamic–pituitary–gonadal (HPG) axis may unmask previously unrecognized ovarian tissue, leading to estrogen excess, gynecomastia, or disruption of HPG‑axis homeostasis (6). Such cyclical activity of residual ovarian tissue has been described but remains insufficiently characterized, particularly in adolescents with fully male external genitalia and no history of undervirilization. Clarifying the dynamic endocrine behavior of residual ovarian components is crucial for optimizing postoperative surveillance and counseling regarding long–term reproductive and endocrine outcomes.

Here, we report an adolescent male with scrotal ovotestes who presented with progressive gynecomastia and displayed cyclical estradiol‑driven fluctuations in gonadotropins and testosterone, corresponding to functional activity of a minute ovarian remnant after testis–preserving partial gonadectomy. This case illustrates the endocrine impact of minimal ovarian tissue during adolescence and highlights practical considerations for surgical decision‑making and long‑term follow‑up in ovotesticular DSD.

Case Presentation

A 13-yr-10-mo-old boy was referred for evaluation of progressive gynecomastia (Fig. 1a). He had been reared as male, and his external genital phenotype was fully male (Figs. 1b, 1c). There was no history of undervirilization, hypospadias, or cryptorchidism. On physical examination, his height was 158.3 cm (−0.48 SD), and his weight was 49.5 kg (−0.14 SD). Pubertal staging revealed Tanner stage B4 for breast development (glandular tissue: right 29 mm, left 16 mm) and Tanner stage G4 for external genitalia (Figs. 1a–1c). Both testes were firm, scrotal, and estimated to be approximately 4 mL in volume by palpation; however, scrotal ultrasonography demonstrated reduced testicular parenchymal volumes (right 0.93 mL, left 1.17 mL). Scrotal ultrasonography further revealed multiple intratesticular cystic lesions, more numerous on the left side. Pelvic ultrasonography identified a cystic lesion posterior to the bladder, which was also confirmed on magnetic resonance imaging (MRI) performed using a 1.5‑T scanner (Fig. 1d). Pelvic MRI demonstrated a small prostate (volume 3 mL) and bilateral intratesticular cysts (three on the left and one on the right) (Figs. 1d–1f). On diffusion-weighted imaging (single-shot spin-echo echo-planar imaging; TR/TE 3000/67.2 ms; b = 0 and 800 s/mm²; slice thickness 3.5 mm without gap; matrix 160 × 64; field of view 350 × 170 mm), apparent diffusion coefficient (ADC) values measured in visually normal testicular parenchyma were 1.72 ± 0.09 × 10⁻³ mm²/s in the left testis and 1.71 ± 0.16 × 10⁻³ mm²/s in the right testis.

Fig. 1.

Clinical and pelvic/gonadal MRI findings. (a) Tanner stage B4 breast development. (b) The urethral meatus opens at the tip of the glans penis. (c) Both gonads are located within the scrotum, and the external genitalia show a completely male phenotype. (d) Pelvic T2‑weighted sagittal MRI demonstrating a large male‑type vagina (*) posterior to the bladder. The prostate is hypoplastic. A cystic lesion with higher signal intensity than the testes is seen within the scrotum (white arrowhead). (e) Preoperative T2‑weighted sagittal and coronal MRI of the right gonad showing a gonad with signal intensity comparable to a normal testis, with a small cystic lesion of higher signal intensity located at the upper pole (black arrowhead). (f) Preoperative T2‑weighted sagittal and coronal MRI of the left gonad showing a testis‑equivalent gonad surrounded by multiple cystic lesions with higher signal intensity than the testis (white arrowhead). (g) Postoperative T2‑weighted sagittal and coronal MRI of the right gonad showing a gonad with normal testicular signal; the previously observed high‑intensity cystic lesion is no longer present. (h) Postoperative T2‑weighted sagittal and coronal MRI of the left gonad showing a gonad with testis‑equivalent signal and one residual high‑intensity cystic lesion.

Baseline biochemical evaluation revealed hypergonadotropic findings with relative estrogen excess: luteinizing hormone 11.6 IU/L, follicle-stimulating hormone 33.6 IU/L, testosterone 3.93 ng/mL (age-appropriate male reference range 1.31–8.71 ng/mL), estradiol (E2) 58.7 pg/mL (elevated for boys aged 14–15 yr; reference range 5–40 pg/mL), progesterone 0.28 ng/mL, and prolactin 10.6 ng/mL. Cytogenetic analysis revealed chromosome mosaicism, 47,XX,+der(8)[14]/46,XX[6], with a derivative chromosome 8, and fluorescence in situ hybridization for SRY was negative. Array comparative genomic hybridization further demonstrated duplications involving pericentromeric regions of chromosome 8 on both the short and long arms. Targeted next-generation sequencing (NGS) using a DSD gene panel (SureSelect) containing AKR1C2, AKR1C4, AR, CBX2, CYP11A1, CYP17A1, DHH, FGF8, HSD3B2, LHCGR, MAMLD1, MAP3K1, MCM4, NR5A1, POR, RBM28, RSPO1, SOX9, SRD5A2, STAR, SRY, TSPYL1, WDR11, WNT4, WT1, and ZFPM2, combined with Sanger sequencing of selected regions identified by NGS for quality control to exclude potential artifacts, identified no pathogenic variants.

At a multidisciplinary conference on DSD, the following issues were discussed: (i) gynecomastia potentially caused by estrogen production from suspected ovarian tissue; (ii) maintenance of the male sex of rearing; and (iii) surgical management of the gonads, including complete excision versus partial resection of the discordant gonadal tissue (ovarian tissue in this case).

After comprehensive counseling regarding endocrine function, fertility potential, and surgical options, written informed consent was obtained from the patient and his family. Endoscopic evaluation of the internal genitalia using laparoscopy and cystoscopy was planned. A testis–preserving surgical approach was selected, with resection of any ovarian tissue if identified, to minimize endocrine disruption while preserving testicular tissue whenever possible.

Cystoscopy revealed a male urethra with a prominent verumontanum and a prostatic utricle, without evidence of a uterus. Laparoscopy demonstrated vas deferens and spermatic vessels entering the internal inguinal rings bilaterally, with no Müllerian structures identified, consistent with a male internal phenotype. Given the presence of firm, whitish foci within both testes, suspicious for ovarian tissue, partial gonadectomy was performed. Intraoperative frozen-section analysis confirmed mixed gonadal tissue, and macroscopically demarcated firm areas were partially excised while preserving testicular tissue bilaterally (Figs. 2a–2c).

Fig. 2.

Intraoperative and histopathological findings. (a) Bilateral ovotestes were identified intraoperatively; firm, whitish areas were suspected to represent ovarian components. (b) Partial resection of the ovarian portions was performed: right ovarian tissue (white arrowhead) and left ovarian tissue (black arrowhead), both confirmed as ovarian tissue by intraoperative frozen section. (c) The tunica albuginea was closed with absorbable sutures. (d) Seminiferous tubules containing only Sertoli cells (white arrowhead) (×200). (e) Seminiferous tubules containing only Sertoli cells, with calretinin‑positive Leydig cells (white arrowhead) (×200). (f) Seminiferous tubules containing only Sertoli cells, negative for Oct3/4 (×200). (g) Epididymal duct lined by pseudostratified columnar epithelium with prominent stereocilia (black arrowhead) (×200). (h) Rete testis (×100). (i) Graafian follicle (black arrowhead) (×200).

Histopathological examination confirmed bilateral ovotestes, with ovarian tissue containing follicles ranging from primordial to Graafian stages, adjacent to testicular tissue composed of seminiferous tubules and interstitial Leydig cells. The seminiferous tubules contained only Sertoli cells without evidence of spermatogenesis (Johnsen score, 3). Rete testis and epididymal tissue were present. Calretinin was positive in Leydig cells, and no OCT3/4‑positive germ cells were identified, arguing against intratubular germ cell neoplasia (Figs. 2d–2i).

Postoperatively, serum E2 levels decreased from 58.7 pg/mL to 12 pg/mL at 0.1 yr, followed by fluctuations (46 pg/mL at 0.2 yr, 30 pg/mL at 0.5 yr, 79 pg/mL at 0.7 yr, and 14 pg/mL at 1 yr 4 mo), paralleling oscillations in testosterone levels (range approximately 0.91–3.38 ng/mL). These changes corresponded with variable volumes of a residual left ovarian component (ovarian remnant), measuring 0.04–0.83 mL on serial imaging (Fig. 3). The patient remained asymptomatic apart from gynecomastia. At 14 yr and 1 mo of age (height 158.5 cm, weight 49.9 kg), bone age was assessed as 15.6 yr using male standards with a corresponding height SDS of −1.73.

Fig. 3.

Longitudinal endocrine profile and imaging-derived ovarian remnant volumes before and after partial gonadectomy. Estradiol (E2, red line, left y axis) and testosterone (T, blue line, right y axis) plotted against age, with green circles indicating left “ovarian remnant” volumes (mL) derived from serial ultrasonography/MRI. The vertical dashed line marks the date of partial gonadectomy (age ~14 yr). An inset ultrasound image illustrates a representative cystic lesion. Serial gonadotropins (FSH, solid black; LH, dashed black). Note the immediate postoperative fall in E2 with subsequent fluctuations paralleling residual ovarian tissue volumes, while testosterone levels remain within the expected pubertal/young‑adult range. E2, estradiol; T, testosterone; FSH, follicle-stimulating hormone; LH, luteinizing hormone.

At 15.6 yr of age, follow-up MRI demonstrated a right testis measuring 3 mL without cysts and a left testis measuring 3 mL with a small ovarian remnant (0.77 mL) (Figs. 1g, 1h). ADC values in the testicular parenchyma remained stable compared with baseline (left 1.66 ± 0.19 × 10⁻³ mm²/s; right 1.66 ± 0.22 × 10⁻³ mm²/s). Lower urinary tract monitoring related to the prostatic utricle, including uroflowmetry, post-void residual measurement, and urinalysis, remained normal. Throughout adolescence, the patient participated in competitive sports, including track and sprinting in junior high school and tennis in high school.

At 18 yr of age, dual-energy X-ray absorptiometry demonstrated normal bone mineral density (areal BMD 1.209 g/cm²). Both gonads remained scrotal without evidence of atrophy. The patient was counseled that testosterone replacement therapy might be required in adulthood. Given his asymptomatic clinical course, improvement of gynecomastia, and fluctuating hormone levels and gonadal volumes possibly related to residual ovarian tissue, definitive excision of the ovarian remnant was discussed but declined by the patient and his family.

Statement of ethics

This study was conducted in compliance with the principles of the World Medical Association Declaration of Helsinki. Ethical approval was not required for this study in accordance with local and national guidelines.

Informed consent

Written informed consent was obtained from the patient’s legal guardians for publication of the clinical details, including images, in this manuscript.

Discussion

Principles of management

Ovotesticular DSD is typically managed to achieve anatomic congruence to the sex of rearing, most commonly through removal of internal structures incongruent with the chosen sex and reconstruction of external genital anatomy, while balancing long-term endocrine and oncologic considerations. Within this framework, conservative gonadal surgery—preserving gonadal tissue concordant with the sex of rearing—is generally preferred (3). In male-reared patients with scrotal ovotestes, excision of the ovarian component with preservation of testicular tissue has been reported to maintain endogenous androgen production, providing an organ-preserving alternative to complete gonadectomy (2, 7). Although a chromosomal abnormality involving chromosome 8 was identified in this case, a causal relationship with ovotesticular DSD remains unclear. Trisomy 8 and related chromosomal duplications are not typically associated with abnormalities of external genital development, and to our knowledge, similar cases have not been reported. Therefore, the genetic findings should be interpreted with caution.

Intraoperative identification challenges

When the interface between ovarian and testicular components is clearly demarcated, testis-preserving resection is typically straightforward; however, large histopathologic series have shown that many ovotestes exhibit admixed or compartmentalised arrangements rather than a simple bipolar pattern (5), complicating intraoperative discrimination of residual ovarian elements. Indistinct planes and small, deeply embedded ovarian foci therefore increase the likelihood of incomplete excision. Intraoperative ultrasonography has been reported to assist in localizing cystic ovarian elements, particularly when the gross appearance is equivocal (8). In our case, the macroscopic boundary was indistinct; we excised regions that felt firm on palpation as presumed ovarian tissue, but non‑palpable microfoci could not be reliably identified, leaving the possibility of residual ovarian tissue. Because only partial gonadectomy was performed, the overall architectural relationship within the ovotestis could not be definitively evaluated, and it remains unclear whether the ovarian and testicular components were clearly separated or exhibited an interdigitating interface.

Cyclical ovarian activity and a “tug of war” on the HPG axis

Postoperatively, we documented intermittent rises in E2. These were accompanied by oscillations in gonadotropins and T, paralleling small volume changes of a left ovarian remnant on serial imaging (Fig. 3). This pattern is consistent with cyclical functional activity of minimal residual ovarian tissue during adolescence. It illustrates how even tiny ovarian remnants can modulate the hypothalamic–pituitary–gonadal axis. Importantly, although testosterone and estradiol may appear to fluctuate in parallel at certain time points, longitudinal evaluation of the endocrine profile reveals a more complex and dynamic pattern. This pattern became evident only through extended follow-up and integrated longitudinal assessment of hormonal dynamics. Such findings highlight the limitations of cross-sectional interpretation. The data suggest that the relationship among estradiol, testosterone, and gonadotropins reflects a dynamic feedback-driven interplay rather than a simple parallel or reciprocal relationship. We conceptualize this as a competing and compensatory “hormonal tug-of-war” within the hypothalamic–pituitary–gonadal axis. Clinically, these hormonal fluctuations corresponded to waxing and waning gynecomastia. Meanwhile, testicular diffusion metrics on MRI and overall pubertal progression remained stable (Fig. 2). The right testicular ADC values were 1.71 ± 0.16 × 10⁻³ mm²/s preoperatively and 1.66 ± 0.22 × 10⁻³ mm²/s postoperatively. This range is comparable to that typically observed in nonobstructive azoospermia (9). Such elevated ADC values are consistent with seminiferous tubular atrophy, thinning of the tubular epithelium, and reduced cellular density. All of these factors increase water diffusivity within the testicular parenchyma. Normal adult ADC values range from about 1.08 to 1.32 × 10⁻³ mm²/s and tend to increase with age. This reflects age‑related decreases in seminiferous tubular diameter and cellularity (10). In our case, histopathology demonstrated a Sertoli‑cell‑only pattern without evidence of spermatogenesis. This corroborates the markedly reduced germ‑cell density implied by the elevated ADC values. These findings suggest that persistently high ADC values reflect underlying structural impairment rather than transient pubertal hormonal variability.

Cyclical ovarian activity within the ovotestes has been described previously. Greeley et al. showed that testicular tissue can suppress ovarian activity in the neonatal period. Removal of the ovotestis may precipitate rapid ovarian hyperfunction during mini-puberty (11). In addition, an ovarian component not identified on neonatal gonadal biopsy may become clinically evident only in adolescence. It can manifest with gynecomastia and elevated E2, together with T fluctuations that stabilize after excision of the ovarian portion. This indicates that even small, unrecognized ovarian remnants can exert clinically meaningful endocrine effects within an ovotestis (6). More broadly, the coexistence of ovarian and testicular tissue from puberty into adulthood may allow E2-driven modulation of the hypothalamic–pituitary–gonadal axis. This can lead to secondary fluctuations in testosterone secretion, as described in previous clinical reports (2, 12, 13). There are reports of phenotypic males with ovotesticular disorder of sex development who lived through childhood and puberty without exogenous hormone replacement, despite limited endogenous testosterone production (12). These oscillatory endocrine dynamics may be disadvantageous for individuals reared as males in maintaining a stable androgenic milieu.

Tissue biology and functional asymmetry

A commonly observed asymmetry in ovotesticular DSD is that testicular elements tend to be hypoplastic or dysgenetic with age, whereas ovarian elements may retain function even at small volumes (3); however, this pattern has been described primarily in cohorts predominantly composed of individuals with a 46,XX karyotype, and the degree and nature of this asymmetry may vary depending on the underlying karyotype and genetic background. Long-term follow-up studies have suggested that, in such 46,XX-dominant cohorts, testicular tissue often shows progressive structural and functional decline, whereas ovarian tissue may remain hormonally active even when limited in volume (3). Our case exemplifies this functional imbalance: a minute ovarian remnant produced sufficient E2 to influence gonadotropins and testosterone, yet the patient maintained day-to-day well-being and typical activities, with normal bone mineral density in late adolescence—an outcome that may be partly explained by intermittent exposure to estradiol. Because bone mineral density represents an integrated, long-term marker of sex steroid action, particularly E2, in the male skeleton (14), the preserved bone mass in our patient is consistent with a potential contribution of estradiol derived from the ovarian remnant, although other sources, such as peripheral aromatization, cannot be excluded.

Oncologic risks and surveillance

The overall risk of malignancy in ovotesticular DSD varies by karyotype, gonadal histology, and the presence of high-risk markers (14). In this patient, OCT3/4 immunostaining was negative, arguing against intratubular germ cell neoplasia at the time of surgery and supporting a testis-preserving strategy. Nonetheless, oncologic risk cannot be dismissed entirely. Current clinical guidelines recommend risk stratification based on karyotype, gonadal location, and histopathological features, with surveillance considered an acceptable strategy in selected low-risk cases (15). In particular, ovotesticular DSD with a 46,XX karyotype and without evidence of germ cell neoplasia is generally regarded as a lower-risk category.

Endocrine and structural monitoring.

A structured surveillance approach that integrates endocrine monitoring with targeted imaging may be useful to detect functional ovarian remnants and assess for interval structural changes, particularly in settings with a relatively low risk of malignancy, where immediate prophylactic gonadectomy is not indicated. Such an approach is not intended for the direct detection of malignant transformation, but rather to identify clinically relevant changes that may warrant further investigation.

Conclusion

Even a small ovarian remnant in an adolescent male with scrotal ovotestes can intermittently produce estradiol, creating short‑term “tug‑of‑war” oscillations between estrogen and testosterone that influence biochemical profiles and gynecomastia. Because the long‑term impact of this hormonal imbalance in male‑reared individuals remains unclear, a testis‑preserving approach should be accompanied by structured, long‑term follow‑up with hormone‑based monitoring and targeted imaging.

Conflict of interests

The authors have no conflicts of interest to declare.

Data availability statement

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.