Primitive ovarian carcinosarcoma: a clinical and radiological analysis of five cases

Abstract

Abstract Background Ovarian carcinosarcomas (OCS) are very rare tumors composed of a mixture of carcinomatous and sarcomatous elements. There have been only scattered case studies that have described the imaging findings. In order to improve the awareness of this rare tumor, this study aimed to analyze the clinical and imaging features of five cases of OCS confirmed by surgical pathologic evaluation. Methods This retrospective study includes five OCS patients diagnosed and treated at our institute. The clinical course and imaging findings of all patients were retrospectively analyzed. The patients were 31 to 59 years of age. All five patients underwent CT scans, two underwent MRI scans. Results The five patients have no specific symptoms. Four patients had elevated CA 125 levels and three patients had elevated CA 153 levels. All patients had unilateral tumors, four in the left ovary, one in the right ovary. The largest transverse diameter of the tumors ranged from 11 cm to 14 cm. Two tumors showed solid masses with unequally sized cystic areas or necrosis, one showed a multilocular cystic mass with a large solid protrusion, two tumors showed a larger cystic mass with multiple mural nodules. The solid components of the tumors demonstrated restricted diffusion (the average ADC value being 998 mm2/s and 1102 mm2/s, respectively), and showed moderate or obvious enhancement. All five patients were treated by surgical resection and adjuvant chemotherapy. One patient is currently undergoing post-operative chemotherapy 1 month after operation and clinical stable. Three patients survived and showed no obvious recurrence and / or metastasis in follow-up from 9 to 59 months. One patient died from recurrence and metastasis. Conclusions OCS are rare and demonstrate variable CT and MRI morphological appearances. Due to the heterogeneous nature and very low morbidity of OCS, combination of careful analysis of imaging findings and clinical features might be useful for a more accurate diagnosis of OCS.