Multimodality Imaging Approach to Ovarian Neoplasms with Pathologic Correlation
Radiographics . Jan-Feb 2021;41(1):289-315. doi: 10.1148/rg.2021200086. Epub 2020 Nov 13.
Erin C Taylor, Lina Irshaid, Mahan Mathur
Ovarian neoplasms can be categorized on the basis of histopathologic features into epithelial surface cell tumors, germ cell tumors, sex cord-stromal tumors, and metastases. While their imaging appearance is often nonspecific, it closely parallels the gross pathologic appearance, and radiologic-pathologic correlation is helpful to aid in a deeper understanding of the subtypes. Epithelial cell neoplasms are the most common category, and they can be benign, borderline, or malignant. Specific subtypes include serous (most common), mucinous, seromucinous, endometrioid, clear cell, Brenner, and undifferentiated. High-grade serous cystadenocarcinoma accounts for the majority of malignant ovarian tumors and the most ovarian cancer deaths. While serous neoplasms are often unilocular and bilateral, mucinous neoplasms are larger, unilateral, and multilocular. Solid components, thickened septa, and papillary projections, particularly with vascularity, indicate borderline or malignant varieties. Endometrioid and clear cell carcinomas can arise within endometriomas. Fibrous tumors (cystadenofibroma, adenofibroma, fibroma or fibrothecoma, and Brenner tumors) demonstrate low T2-weighted signal intensity of their solid components, while teratomas contain lipid. The nonspecific imaging appearance of additional malignant ovarian germ cell tumors can be narrowed with tumor marker profiles. Sex cord-stromal tumors are often solid, and secondary signs from their hormonal secretion can be a clue to their diagnosis. The authors review the anatomy of the ovary and distal fallopian tube, the proposed origins of the histologic subtypes of tumors, the clinical features and epidemiology of ovarian neoplasms, and the applications of US, CT, and MRI in imaging ovarian neoplasms. The main focus is on the radiologic and pathologic features of the multiple ovarian neoplasm subtypes. An algorithmic approach to the diagnosis of ovarian neoplasms is presented.
Read Full Article Here: https://doi.org/10.1148/rg.2021200086