Frozen section is a reliable technique in gynecologic pathology and is widely used to guide intraoperative management in patients presenting with ovarian masses. However, there are limited data regarding the diagnostic accuracy of frozen section in specific subtypes of ovarian neoplasia. Our impression that primary clear cell carcinoma (CCC) causes disproportionate diagnostic difficulty led us to review the intraoperative and final histopathologic reports from a consecutive series of 44 CCC that were subject to frozen-section assessment and to compare the results with a similar number of primary serous and endometrioid carcinomas. The original intraoperative slides from those CCC with discordant diagnoses were also reviewed. Review of the diagnostic reports showed that CCC was less frequently specifically identified than serous or endometrioid carcinomas on frozen section (44% cases compared with 55% and 65%, respectively), although the differences were not statistically significant. Difficulties in distinguishing primary ovarian carcinoma from tumors metastatic to the ovary occurred in a minority of cases of all histologic subtypes, but was slightly more frequent in CCC. Two CCC were misdiagnosed as borderline epithelial tumors and 1 case as a dysgerminoma. Review of the frozen-section slides from the CCC with discrepant intraoperative diagnoses showed features suggestive or indicative of the correct diagnosis in 7 (39%) of 18 cases.
Primary endometrial serous carcinoma, known for its aggressive nature and poor prognosis, shares similarities with breast and gastric cancers in terms of potential HER2 overexpression as a therapeutic target. Assessing HER expression is complicated by tumor heterogeneity and discrepancies between primary and metastatic sites. In this study, we retrospectively analyzed HER amplification and expression in 16 pairs of primary endometrial serous carcinoma resections and corresponding metastases. HER2 status was determined using immunohistochemistry (IHC), with criteria based on the percentage and intensity of tumor cell staining. Confirmatory techniques, such as dual in situ hybridization (DISH) and fluorescence in situ hybridization (FISH), were also employed. This study reports on the concordance rates and the presence and pattern of HER2 heterogeneity. Our results showed an 87.5% concordance rate in HER2 amplification status between primary and metastatic sites, with 33% of cases scored as 2+ being amplified. Heterogeneity was observed in 100% of amplified cases and 95% of non-amplified cases on in situ testing, with variations in heterogeneity patterns between techniques. In conclusion, our findings emphasize the importance of testing both primary and metastatic sites or recurrences, with a concordance rate of 87.5%. In addition, a review of the literature and combining the results showed a concordance rate of up to 68%. The presence and pattern of heterogeneity, particularly in cases of mosaic or clustered heterogeneity in the primary tumor, may serve as reliable indicators of concordance, predicting a non-amplified HER2 status in corresponding metastases.