1. Introduction

Ovarian sarcomas are rare, comprising only 2–3% of ovarian tumours [1] and with an incidence of 0.53 per 100,000 [2]. Their prevalence is higher in post-menopausal women [2, 3, 4] and they often present with non-specific symptoms such as abdominal pain and bloating [3, 5, 6]. Ovarian sarcomas are known to be highly malignant and aggressive, and frequently present in advanced stages of disease [2, 7]. This results in a poor prognosis, with median survival of 16–18 months [3, 7], five-year survival rates of 4–24% [2, 7] and a poorer prognosis than both their uterine counterparts [6] and epithelial ovarian carcinoma across all disease stages [8].

Sarcomas originate from mesenchymal cells and can be classified by cell morphology and tissue type. Ovarian sarcomas are a heterologous group of tumours. Histological subtypes include leiomyosarcoma, endometrial stromal sarcoma, fibrosarcoma, rhabdomyosarcoma and adenosarcoma [9]. Adenosarcomas are characterised by both benign epithelial and malignant mesenchymal components [4]. The mesenchymal component is most commonly low-grade endometrial stromal sarcoma but occasionally exhibits high-grade undifferentiated sarcomatous overgrowth [10]. They may also occasionally contain heterologous stromal elements or sex cord-like differentiation [10]. Spindle cell components have been described in many different ovarian lesions, both neoplastic and non-neoplastic [11], including a small number of reported spindle cell sarcomas in the ovary [12, 13, 14]. Because of the rarity and heterogeneity of ovarian sarcomas, they are often studied as a group and only reported in small case series. Consequentially, our understanding of their behaviour is limited and there are no evidence-based guidelines on their management.

DICER1 mutations are an emerging finding in sarcomas [15, 16, 17, 18]. DICER1 is a gene that encodes a ribonuclease III endonuclease involved in the biogenesis of microRNAs. MicroRNAs regulate messenger RNAs, which regulate gene expression [16]. Dysregulation by DICER1 mutations adversely impacts their tumour suppressor function [16]. Germline mutations in DICER1 are associated with an increased risk of a range of tumours, most notably pleuropulmonary blastomas, thyroid lesions and cystic nephromas [16, 19]. This predisposition is termed DICER1 syndrome. In the gynaecological tract, germline DICER1 mutations are associated with ovarian sex cord-stromal tumours—mainly Sertoli-Leydig cell tumours—and embryonal rhabdomyosarcomas of the cervix [17, 18]. DICER1 mutations have also been identified in gynaecological adenosarcomas. The vast majority of these are somatic and identified in tumours arising from the uterus [15, 16, 17, 18, 20, 21, 22, 23]; there are only scarce examples of uterine adenosarcomas associated with DICER1 syndrome [23, 24]. In ovarian sarcomas, there are rare reports of DICER1 mutations. To our knowledge, there are only two previously reported cases of ovarian adenosarcomas harbouring DICER1 mutations, both of which occurred in the paediatric population [16, 25]. The first was an ovarian sarcoma in a 10-year-old female who had both somatic and germline DICER1 mutations and subsequently developed a Sertoli-Leydig cell tumour [25]. The second was a high-grade pleomorphic sarcoma with two DICER1 variants, determined to be germline in origin and with a significant family history of malignancy, in a 5-year-old female [16]. Additionally, the case of a metastatic peritoneal adenosarcoma in a 16-year-old with two presumed somatic DICER1 mutations has been reported [25].

We present the case of a 35-year-old with ovarian adenosarcoma with rhabdosarcomatous overgrowth and DICER1 mutations. To our knowledge this is the first reported case of an ovarian adenosarcoma with DICER1 mutations in the adult population.

2. Case presentation

A 35-year-old nulliparous woman presented in January 2024 with two days of abdominal pain, fevers, nausea, and diarrhoea. There was no abnormal vaginal bleeding or discharge, and bowel and bladder function were preserved. The woman had an unremarkable medical and family history and no sexual risk factors. She had recently travelled overseas to South-East Asia two months prior.

On examination, her abdomen was mildly distended with generalised tenderness, worst in the right lower quadrant. Laboratory investigations were most remarkable for elevated inflammatory markers (white cell count 10.3 × 10⁹/L, C-reactive protein 186 mg/L.) A computed tomography (CT) scan demonstrated a large solid/cystic mass in the right adnexa measuring 13 × 9 × 10 cm with ascites (Fig. 1). A pelvic ultrasound characterised the lesion as hyperechoic and avascular. Given the septic symptoms, a tubo-ovarian abscess was favoured as the provisional diagnosis, so the patient was treated with intravenous antibiotics. CT-guided drainage was attempted but unsuccessful; instead only ascitic fluid was able to be drained and was found to contain an inflammatory cell infiltrate. Cytology of the fluid was negative. Extensive investigations for an infectious source were negative, including cultures of blood, urine and stool, sexually transmitted infection screen, and serology for human immunodeficiency virus (HIV), syphilis, entamoeba histolytica and tuberculosis.

Fig. 1.CT and ultrasound images of the pelvic mass. (A,B) CT contrast-enhanced coronal (A) and axial (B) planes demonstrating the large right pelvic mass (white arrows) which is predominantly cystic, with internal solid enhancing component. The mass had rapidly increased to 16 × 11 × 14 cm from 13 × 9 × 10 cm from 8 days prior. There is ascites (white arrow heads). (C) Pelvic ultrasound showing a hyperechoic mass that is avascular. A normal right ovary was not identified.

Despite antibiotics, the patient continued to have daily fevers and repeat CT (at a one-week interval) demonstrated growth of the mass to 16 × 11 × 14 cm and increased ascites. Pelvic magnetic resonance imaging (MRI) was performed, which showed a multiloculated solid/cystic lesion with large papillary projections, and loculated pockets of ascites in the pelvis and paracolic gutters (Fig. 2). The ascites had dependent fluid-fluid levels and layering T2 hypointensity, in keeping with the presence of blood products. There were no omental nodules. These features were highly suspicious for a primary ovarian neoplasm with recent lesion rupture. Tumour markers were performed, finding an elevation of cancer antigen (CA) 125 at 383 U/mL. CA15.3, CA19.9, carcinoembryonic antigen (CEA), alpha-fetoprotein (AFP), and lactate dehydrogenase (LDH) were within normal limits. The patient was then referred to our tertiary gynae-oncological institution for further evaluation.

Fig. 2.MRI images of the pelvic mass. (A) MRI coronal contrast-enhanced T1 fat suppressed sequence demonstrating heterogeneous enhancement of the solid/cystic right pelvic mass (white arrows). (B,C) MRI axial and sagittal T2 sequences. There is loculated ascites, which is complex with thick septations and dependent fluid-fluid levels in the vesicouterine and rectouterine pouches, as well as in the left paracolic gutter (black arrows). The signal is in keeping with layering haemorrhage. U (uterus), R (rectum), B (bladder), Mass (right pelvic mass).

Upon review at our institution the patient had deteriorated, with worsening pain, significant abdominal swelling, vomiting, anorexia, and breathlessness. Her abdomen was grossly distended with shifting dullness, and she demonstrated signs of peripheral fluid overload. Inflammatory markers worsened and she became anaemic and hypoalbuminaemic. She developed an oxygen requirement and subsequent CT pulmonary angiogram demonstrated bilateral pleural effusions. Thoracentesis was performed and a peritoneal drain was inserted, draining 5 L of fluid in total. Both ascitic and pleural fluids demonstrated acute inflammatory cell infiltrates; no organisms were grown, and no malignant cells were seen.

The patient proceeded to diagnostic laparoscopy. A large pelvic mass, inflammatory adhesions to the abdominal wall, haemoserous ascites and evidence of pre-operative rupture were identified. The operation was converted to laparotomy and a 15 cm mass with solid and cystic components was found encapsulating the right ovary. A right salpingo-oophorectomy was performed; the mass was friable and therefore removed piecemeal. The left ovary appeared normal, the left fallopian tube was oedematous, the uterus was erythematous and there was inflammatory stranding in the right upper quadrant, but no evidence of macroscopic extra-ovarian disease was found. Abdominal washout and omental biopsy were performed.

Macroscopically, the ruptured cystic mass had a largely haemorrhagic soft texture with areas of purulent exudate. Microscopically, the tumour was a biphasic neoplasm with areas of glandular and mesenchymal differentiation. It contained pleomorphic spindle cells with a high mitotic rate and significant nuclear overlapping, diffuse areas of geographical necrosis, stromal atypia, and a brisk stromal mitotic rate, most noticeable in areas of sarcomatous overgrowth, and heterologous elements including rhabdoid and chondroid differentiation (Fig. 3). There were distinctive areas of periglandular condensation of endometrial type stroma, also termed a “cambium layer”, suggestive of adenosarcoma. No Leydig cells were identified. There was no evidence of origin in endometriosis. Immunostaining was positive for cluster of differentiation (CD) 10, focally positive for myogenin, myogenic differentiation 1 (myoD1), desmin and CD56, and negative for inhibin, calretinin and steroidogenic factor 1 (SF1) (See Supplementary Fig. 1). The omental biopsy was negative. Comprehensive genomic profiling of the tumour via next-generation sequencing identified three somatic mutations of potential clinical significance in tumour suppressor genes DICER1 (D1709N c.5125G>A, 42% and Q1389* c.4165C>T, 38%) and NOTCH3 (c.1607-1G>T).

Fig. 3.Pathology images. (A) Macroscopic image of specimen removed at the time of surgery. (B) Malignant spindle cell neoplasm, ×100 magnification. (C) Tumour cambium layer, ×100 magnification. (D) Heterologous cartilage, ×40 magnification.

Differential diagnoses considered due to the underlying biphasic nature of the tumour included carcinosarcoma with heterologous elements, however the glandular component was bland endometrial type epithelium with no significant atypia [6]. A poorly differentiated Sertoli-Leydig cell tumour with sarcomatoid features was also considered. This uncommon entity may show background evidence of solid tubules, hollow tubules, Leydig cells, or retiform growth after diligent searching [26, 27], and would usually demonstrate some positive staining with inhibin, calretinin and SF1. A cambium layer is not a usual characteristic of these lesions however there can be significant morphological overlap between cases. Hence, the conclusion of an adenosarcoma with sarcomatous overgrowth and DICER1 mutations was reached.

The patient was referred to Medical Oncology and the Pelvic and Retroperitoneal Sarcoma multidisciplinary team at our institution. Formal staging was completed with positron emission tomography (PET), demonstrating subsequent metastatic spread to the left upper quadrant, midline upper abdomen and left posterior pleura. Further surgery and radiotherapy were not recommended. The patient declined germline testing for the identified DICER1 mutations, as she does not have any siblings or children and is not planning to conceive in the future. Unfortunately, none of the identified genetic variants in this case were therapeutically actionable following review by our molecular oncology board. She received palliative-intent chemotherapy commencing in March 2024 with six cycles of doxorubicin followed by maintenance chemotherapy with oral cyclophosphamide. This was associated with a complete radiologic and metabolic response that was maintained for 14 months from commencement of chemotherapy. Unfortunately, there was subsequent rapid and large volume recurrence in May 2025. She underwent debulking surgery due to a resultant bowel obstruction, where 1411 g of tumour was removed from the abdominopelvic cavity. Her prognosis remains guarded.

3. Discussion

This is a rare case of a poorly differentiated high grade ovarian adenosarcoma with DICER1 mutations in a young woman. It is a unique presentation and entails a rare histopathological and molecular classification that presented diagnostic and management challenges. Additionally, it is the only reported case, to our knowledge, of an ovarian adenosarcoma containing DICER1 mutations in an adult.

The presented tumour was unilateral, 15 cm in maximal dimension, contained solid and cystic components and its histopathological features, immunohistochemistry and molecular profiling led to the favoured diagnosis of adenosarcoma with sarcomatous overgrowth. The tumour has both similarities and differences compared to previously reported ovarian sarcomas. Similar features include its large size, unilateral presence, presenting symptoms, advanced disease stage at diagnosis, recurrence, and poor prognosis. In a previous series of 40 cases of ovarian adenosarcoma [6] the tumours were unilateral in 97.5% of cases, ranged in size from 5–50 cm with a mean largest diameter of 14 cm, presented most commonly with symptoms of abdominal discomfort and distension, and recurred in 77% of cases with sites of recurrence being intra-abdominal and lung metastases. In another series of 47 ovarian sarcomas, noted however of varied histological subtypes, 87% of cases presented in advanced disease stages and 72% developed recurrent disease despite post-operative therapy, with average time to recurrence of 10.5 months [3]. A series of 24 cases of primary ovarian sarcomas, similarly of mixed histological subtypes, were unilateral in 79% of cases, had a mean diameter of 12 cm, and presented in stage III or greater in 92% of cases [7]. Whilst the tumour in our case responded well initially to chemotherapy, its recent large volume recurrence aligns with the known high recurrence risk of extra-uterine adenosarcomas [4] and previously reported poor prognoses [2, 3, 6, 7, 8], particularly of those exhibiting poorly differentiated sarcomatous overgrowth [4]. Immunostaining in our tumour was focally positive for desmin. This is consistent with a previous review of adenosarcoma cases in which two cases demonstrating sarcomatous overgrowth with rhabdomyoblastic differentiation both stained positive for desmin [28]. Negative immunostaining for calretinin is also consistent with previously reported patterns [28].

The tumour, however, also exhibited several uncommon features. Firstly, it presented in a pre-menopausal woman. This is atypical—ovarian sarcomas occur more commonly in post-menopausal women [4] with an incidence up to three times that of pre-menopausal [3]. When compared to the previous cases of DICER1-associated ovarian adenosarcomas, her age is also unusual, with all previously reported cases occurring in children [16, 25]. Secondly, the omentum is usually involved [3, 7], however our patient had a negative omental biopsy. Thirdly, DICER1 mutations were identified, which are a rare but emerging finding in some gynaecological sarcomas [16, 23, 25], and have not been previously identified in an adult ovarian adenosarcoma. Finally, there was evidence of pre-operative rupture, which was found in only 10% of cases in a previous series of ovarian adenosarcomas [6]. Interestingly, in this same case series, features associated with recurrence or extra-ovarian spread were age <53 years, tumour rupture, a high grade, and the presence of high-grade sarcomatous overgrowth [6], all features exhibited in our presented case.

Molecular analysis identified three mutations of clinical significance in tumour suppressor genes DICER1 and NOTCH3. Loss of function NOTCH3 mutations are tumour suppressive in some solid tumours [29], but no link with ovarian sarcomas has been established. DICER1 mutations are a rare but emerging finding in gynaecological adenosarcomas but have mainly been identified in uterine lesions. An analysis of 19 cases of uterine adenosarcoma found somatic DICER1 mutations in eight tumours (42%), half of which demonstrated rhabdomyosarcomatous differentiation [15]. By contrast, Yao et al. [30] analysed 22 cases of adenosarcoma, the majority of which were uterine, and found no DICER1 mutations. Review groups conclude that there is molecular and genetic heterogeneity amongst gynaecological sarcomas, with a rare subset demonstrating DICER1 mutations [17, 30].

Two previous cases of uterine adenosarcoma have harboured an identical DICER1 mutation—D1709N c.5125G > A—to our presented case [15, 18]. The first was a high-grade adenosarcoma with stromal overgrowth and rhabdomyoblastic differentiation in a 35-year-old [15] and the second a high-grade round cell neoplasm also with rhabdomyosarcomatous elements in a 30-year-old [18]. Interestingly, this same mutation is identical to the one previously identified in the aforementioned case of a 10-year-old with an ovarian sarcoma containing DICER1 mutations [25]. Interestingly, she had an additional germline DICER1 mutation and subsequently developed a Sertoli-Leydig cell tumour [25]. The DICER1 mutations reported in the only other case of ovarian adenosarcoma—a five-year-old with a strong family history—were different [16], as were those in the case of the 16-year-old with metastatic peritoneal adenosarcoma [16]. Interestingly, in the latter case, the patient had peritoneal metastases, complex ascites, and bilateral pleural effusions, and an elevated CA125 (464 U/mL), a very similar clinical picture to that of our presented case. She was subsequently found to have a co-existing papillary thyroid microcarcinoma, and whilst germline testing was not performed, was considered suspicious for an underlying DICER1 syndrome [16].

As our patient declined germline testing, it is not possible to say whether the identified DICER1 variants were somatic or germline in origin, although there was no personal or family history of malignancy. It is also possible that the DICER1 mutations may not have been the primary driver of the neoplastic process. However, it is noted that our patient’s tumour exhibited many of the characteristic features of DICER1-associated sarcomas: poorly differentiated spindle cells, areas with rhabdomyoblastic differentiation, a cambium layer, solid and cystic architecture, and foci of cartilage [23]. What remains unclear and debated in the literature is whether the mutations are incidental findings or truly causal, and further research is needed into the role of both somatic and germline DICER1 mutations in adenosarcomas.

Gynaecological sarcomas, as a heterogenous group of varied morphologic, molecular, and clinical manifestations [31], are often initially unrecognised and misdiagnosed [9]. They present a diagnostic challenge to clinicians, likely significantly contributed to by their non-specific presenting symptoms, which are consistently abdominal pain and bloating [3, 5, 6, 7, 12, 13]. A notable feature of our case was the pre-referral delay in diagnosis and evidence of rapid disease progression in the intervening time. This could be partly attributed to the non-specific symptoms of abdominal pain and bloating, as well as the uncommon features of persistent fevers and nausea [6, 7], thus mimicking an infectious process. Radiologically, there is overlap between the features of ovarian neoplasm and tubo-ovarian abscess, which likely clouded the clinical picture. The finding of ascites may have presented an earlier clue, as this is a frequent phenomenon in ovarian sarcomas—79% in one case series [3]. During the period of diagnostic uncertainty, there was significant growth of the tumour—from 13 × 9 × 10 cm to 16 × 11 × 14 cm in one week—and evidence of lesion rupture at surgery, highlighting its aggressive nature. Knowledge of the presentation and aggressiveness of rare ovarian tumours is important in such unusual cases, so they can be considered as a differential diagnosis, potentially leading to more rapid identification in future and improved clinical outcomes.

Our patient underwent a fertility sparing right salpingo-oophorectomy, abdominal washout and omentectomy due to age and fertility status. The patient was disease-free and well at 14 months of follow-up; however, has since developed a large volume recurrence. Whilst her initial response was pleasing given the overall poor prognosis of ovarian sarcomas, recurrence was not an unexpected consequence. The planning and quality of initial surgical resection seems to be a relevant prognostic factor for ovarian sarcomas. In two separate case series, it was the most significant prognostic factor—in the first, the five-year survival rate was 45% compared to 8% with optimal vs. suboptimal surgical cytoreduction [3], and in the second, the one-year survival rate was 71% compared to 29% [7]. This is likely explained by unplanned excision increasing the chance of contamination and residual disease. Other positive predictors of survival are diagnosis at an early disease stage, homologous histological subtypes and pre-operative CA125 level <75 U/mL [3]. It has therefore been proposed that suspected pelvic sarcomas should be biopsied to enable surgical planning and prevent suboptimal debulking [9]. However, this review group admit that in the case of ovarian sarcomas, diagnostic surgery is most often required. This is supported by another review group [4] who conclude that a diagnosis of extrauterine adenosarcoma is unlikely to be known pre-operatively. Our case exemplifies this. Pre-operative attempts were made at gaining a tissue diagnosis, through initial attempted drainage and subsequent ascitic and pleural taps, however these were ultimately fruitless and diagnostic surgery was required. Post-operatively, establishing an accurate histopathological diagnosis is often an additional challenge [6], also exemplified in the thorough and detailed histological and molecular analysis undertaken in our case.

Since ovarian sarcomas are uncommon and heterologous, there are no clear management guidelines. There is consensus, however, that management of rare sarcomas should be conducted in tertiary referral centres with input from a multidisciplinary team of Gynaecological Oncologists, Medical Oncologists, Pathologists and Radiologists [9, 32]. This has been associated with improved overall survival in sarcomas regardless of primary site [32]. Combining accurate histopathological classification, immunohistochemistry, and molecular analysis to create personalised treatments is likely the way forward for patients with rare tumours [31].

4. Conclusion

This is a rare case of advanced ovarian adenosarcoma with DICER1 mutations in a 35-year-old, the first reported of its kind. It adds knowledge to the expanding body of evidence regarding sarcomas exhibiting DICER1 mutations and provides insight into this rare tumour type’s presentation and behaviour. It demonstrates the importance of immunohistochemistry and molecular testing in reaching an accurate diagnosis. Further study is required of rare histological subtypes of sarcoma and their molecular characteristics to expand our knowledge and develop management guidelines, particularly regarding the role of somatic DICER1 mutations in rare sarcoma types.

5. Limitations

The authors note some limitations to this case report. Firstly, since the patient declined germline testing and this was respected by her treating clinicians, it cannot be determined whether the identified DICER1 mutations were acquired or germline in origin. Secondly, given the rarity and heterogeneity of ovarian sarcomas, the data available in the literature is spread across the last several decades. During this time there have been advancements in diagnostic and molecular techniques making comparison of this case to those previously reported challenging, and these comparisons should be interpreted with caution. Finally, the duration of follow-up. This case has been reported 15 months since original presentation; hence longitudinal progress and prognosis beyond this time is not yet available, although there has been a significant recurrence in this time.

Availability of data and materials

The data is contained within the article.

Author contributions

AM—Conceptualisation, Investigation, Writing–original draft, review, and editing. SAS—Supervision, Writing–review & editing. LA—Resources (histopathological images), Writing–review & editing. PG—Writing–review & editing. JY—Resources (radiological images and captions), Writing–review & editing (radiological content only). All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript.

Ethics approval and consent to participate

The research leading to this case report has been approved by the Sydney Local Health District Human Research Ethics Committee (SLHD HREC), under a waiver of consent for sarcoma patients at our institution (Chris O’Brien Lifehouse). The protocol number is X19-0246 and ethics approval number 2019/ETH1837. Additionally, the patient has signed an explicit written consent for both inclusion of her tumour in our institution’s biobank and database, and for publication of this case report and accompanying images. Copies of these written consent forms and the above protocol and ethics approval are available for review by the Editor-in-Chief of this journal on request.

Acknowledgment

Nil additional applicable acknowledgements other than contributing authors.

Funding

This research received no external funding.

Conflict of interest

The authors declare no conflict of interest.

Supplementary material

Supplementary material associated with this article can be found, in the online version, at https://oss.ejgo.net/ files/article/1989228939485036544/attachment/ Supplementary%20material.docx.