Giant primary ovarian fibrosarcoma: Case report and review of pitfalls.

Primary ovarian fibrosarcomas are very uncommon neoplasms; less than one hundred cases have been reported. Diagnosis can be difficult to establish and other similar appearing mesenchymal lesions must be ruled out. A 23 year old nulliparous woman presented with a huge right ovarian tumor (25 × 17 × 12 cm). Ultrasonography guided aspiration was performed and a strong possibility of fibrosarcoma was suggested based on cytomorphological findings. No epithelial element was noted. She underwent surgical excision which confirmed the diagnosis of primary ovarian fibrosarcoma (the largest till date to our knowledge). Immunohistochemical analyses showed vimentin and CD 34 positivity. We speculate that ovarian fibrosarcoma at the preoperative examination might be considered in the differential diagnosis whenever a unilateral, heterogeneous, largely round solid ovarian lesion is detected.

Introduction

Ovarian neoplasms are common with surface epithelial tumors being the commonest overall and germ cell tumors in children. Fibrosarcoma is a rare entity in the ovary; less than one hundred cases have been reported so far. Isolated cases of unilateral primary fibrosarcoma of the ovary[12] and a rare case of bilateral ovarian fibrosarcoma have been reported.[3] It has also been reported in association with the basal cell nevus syndrome[4] and Maffucci syndrome.[5] A case of primary ovarian fibrosarcoma is being presented because of its rarity, giant size and possible cytological and histological diagnostic dilemmas and pitfalls.

Case Report

A 23 year old unmarried nulliparous female without any feminization or masculinization features presented with abdominal distension, intermittent pain abdomen and significant weight loss. Abdominal ultrasonography (USG) revealed a large, solid mass in the right adnexa. The contralateral ovary was normal. No lymph-node enlargement was detected. Other abdominal viscera were unremarkable. Significant ascites was present. Serum α–fetoprotein and chorionic gonadotropin levels were not raised.

USG guided aspiration yielded blood mixed material showing clusters of highly pleomorphic oval to spindle shaped cells with indistinct cell borders [Figure 1a] and atypical mitotic figures. A dissociated cell pattern was also observed at places [Figure 1b] along with few bizarre nuclei [Figure 1c]. No evidence of irregular branching or papillary configuration was seen. Psammoma bodies were absent. No acinar pattern or intracytoplasmic mucin was noted. There was no picket fence appearance as seen in mucinous neoplasms or coffee bean shaped nuclei of granulosa cell tumors. Call-Exner bodies were not found. Absence of abundant clear cytoplasm ruled out dysgerminoma. A cytological diagnosis of primary pleomorphic/ spindle-shaped sarcoma, possibly high grade fibrosarcoma, was proposed. Histological confirmation was suggested after taking adequate sections to rule out any malignant epithelial components missed due to sampling error that may occur in aspirated malignant mixed mesodermal tumors. Peritoneal fluid cytology was non-contributory.

Figure 1

(a) Clusters of pleomorphic spindle cells with indistinct cell borders (Pap, ×100). (b) Oval to spindle shaped cells with bizarre nuclei (arrow) (Pap, ×100)

Following surgical excision, grossly, the right ovary measured 25 × 17 × 12 cm with a smooth external surface [Figure 1d]. Sectioning revealed a solid, firm, grayish mass containing areas of hemorrhage and cystic degeneration. Histologically, the tumor was composed of spindle cells arranged in sheets and intersecting fascicles creating a diffuse herringbone appearance. There was mild to moderate nuclear atypia with enlarged, hyperchromatic, and oblong pleomorphic nuclei with prominent nucleoli. The tumor diffusely involved the entire ovarian stroma with no normal structures remaining. No epithelial component could be identified even after thorough sectioning. A formal mitotic count was carried out on all representative sections of the tumor by recording number of mitotic figures in 50 consecutive high-power fields (HPF) and calculating the average per 10 HPF. The mitotic count varied from 10 to 12 per 10 HPF. These features, particularly the raised mitotic activity, were diagnostic of fibrosarcoma. This was the largest ovarian fibrosarcoma reported till date to the best of our knowledge. Reticulin stain showed fibers separating individual neoplastic cells. Immunohistochemical stains were positive for vimentin and CD34.

Discussion

Primary fibrosarcomas of the ovary are rare and may arise de novo or as a result of malignant change in a pre-existing ovarian fibroma.[6] Guided aspirations are done to plan preoperative treatment protocols. USG guided aspiration in our case showed clusters of highly pleomorphic oval to spindle shaped cells with indistinct cell borders. Atypical mitotic figures along with few bizarre nuclei were also observed. No evidence of irregular branching or papillary configuration observed in serous neoplasms was seen. Psammoma bodies were also absent. No acinar pattern and intracytoplasmic mucin or vacuoles resulting in picket fence appearance encountered in mucinous neoplasms were seen. Coffee bean shaped nuclei and Call-Exner bodies of granulosa cell tumors were not found. Absence of abundant clear cytoplasm ruled out dysgerminoma. A cytological diagnosis of primary pleomorphic/spindle-shaped sarcoma, possibly high grade fibrosarcoma, was proposed considering the morphology, radiological solid appearance, age, absence of feminization or masculinization features with normal serum α–fetoprotein and chorionic gonadotropin levels.

Criteria for the diagnosis of fibrosarcoma of ovary were established by Prat and Scully.[7] They identified mitotic count as the most important feature for distinguishing between benign and malignant lesions. When they separated the tumors into two groups – cellular fibromatous lesions with mitotic counts up to 3 per 10 HPF (11 cases); and those with counts exceeding 3 per 10 HPF (6 cases) – a significant difference in survival was observed. The degree of cytologic atypia was a less reliable prognostic indicator. Although in their study, the lesions selected were pure fibromas showing no evidence of thecomatous differentiation, in a later review, Young and Scully[8] suggested that, in the absence of established criteria for malignancy, the same criteria could be applied to fibrothecomatous lesions. Despite the apparent reliability of mitotic counts in differentiating fibrosarcomas from cellular fibrothecomas, a few examples exist in which cellular fibrothecomatous lesions demonstrating low mitotic activity show an unexpected aggressive behavior.[9] Tsuji et al.[10] established that the MIB-1 (Ki-67) labeling index (LI) in atypical fibromatous lesions of the ovary was reflective of the potential aggressiveness of the tumor. MIB-1 LI for cellular fibromas ranged from 0.5 to 4.0 with a median of 2.3, while that for fibrosarcomas ranged from 3.0 to 10.8 with a median of 6.6. MIB-1 LI was related to the mitotic count, and their results may support a formal mitotic count as a diagnostic tool, especially in cases with a count of three to four mitoses per 10 HPF.

Primary ovarian fibrosarcomas, although rare, must be considered in the differential diagnosis of a unilateral, heterogeneous, solid ovarian lesion. Although usually encountered in the peri/postmenopausal age group, they can occur at any age as seen in our case. They are often underdiagnosed pre-operatively, resulting in diagnostic dilemmas. Cytological suspicion benefits the patient with proper hormonal assessment, intra-operative surgical excision protocols and postoperative management. However, a cautious approach is warranted categorically ruling out other common possibilities as discussed.