Undifferentiated and dedifferentiated endometrial carcinomas (UDEC) are aggressive uterine tumors which may show loss of expression of SMARCA4 (BRG1) or SMARCB1 (INI-1). The recently described SMARCA4-deficient undifferentiated uterine sarcoma (SDUS) has a morphology which overlaps with UDEC. In this study, we compared clinical, morphologic, immunohistochemical, and molecular characteristics to identify features which differentiate SDUS from UDEC. Cases of SDUS (n=12) were compared with cases of UDEC (n=84, 55 of which were previously published). Immunohistochemistry was performed for p53, mismatch repair proteins, claudin-4, SMARCA4, and SMARCB1. Targeted molecular profiling was performed on 15 cases. Patients with SDUS were significantly younger than those with UDEC (mean 35.8 vs. 61.2 y, P=0.0001). UDEC and SDUS showed morphologic overlap; however, phyllodiform architecture favored a diagnosis of SDUS (36% vs. 0%, P=0.005), while prominent nuclear pleomorphism was only seen in some cases of UDEC (0% vs. 24%, P=0.15). Compared with SDUS, UDEC more frequently showed TP53 mutations (0% vs. 34%, P=0.03), microsatellite instability (0% vs. 44%, P=0.006), and intact SMARCA4 and SMARCB1 (0% vs. 80%); a panel combining these immunohistochemical markers had a sensitivity of 100% and specificity of 92% in distinguishing SDUS and UDEC. Cases of UDEC had mutations in genes associated with endometrial adenocarcinomas (eg, TP53, PTEN, PIK3CA) and occasionally SMARCA4, while SDUS was characterized solely by inactivating mutations in SMARCA4. Disease-specific survival was shorter in SDUS than UDEC (median survival 9 and 36 mo, P=0.01). In conclusion, SDUS occurs in younger patients than UDEC, has a worse prognosis, and in most cases has a distinct molecular and immunohistochemical profile.
Women with Lynch syndrome (LS) are at increased risk for the development of epithelial ovarian cancer (OC). Analogous to previous studies on BRCA1/2 mutation carriers, there is evidence to suggest a histotype-specific association in LS-associated OCs (LS-OC). Whereas the diagnosis of high-grade serous carcinoma is an indication for BRCA1/2 germline testing, in contrast, there are no screening guidelines in place for triaging OC patients for LS testing based on histotype. We performed a centralized pathology review of tumor subtype on 20 germline mutation-confirmed LS-OCs, on the basis of morphologic assessment of hematoxylin and eosin-stained slides, with confirmation by immunohistochemistry when necessary. Results from mismatch-repair immunohistochemistry (MMR-IHC) and microsatellite instability (MSI) phenotype status were documented, and detailed pedigrees were analyzed to determine whether previously proposed clinical criteria would have selected these patients for genetic testing. Review of pathology revealed all LS-OCs to be either pure endometrioid carcinoma (14 cases), mixed carcinoma with an endometrioid component (4 cases), or clear cell carcinoma (2 cases). No high-grade or low-grade serous carcinomas or mucinous carcinomas of intestinal type were identified. Tumor-infiltrating lymphocytes were prominent (≥40 per 10 high-powered fields) in 2 cases only. With the exception of 1 case, all tumors tested for MMR-IHC or MSI had an MMR-deficient phenotype. Within this cohort, 50%, 55%, 65%, and 85% of patients would have been selected for genetic workup by Amsterdam II, revised Bethesda Guidelines, SGO 10% to 25%, and SGO 5% to 10% criteria, respectively, with <60% of index or sentinel cases detected by any of these schemas. To further support a subtype-driven screening strategy, MMR-IHC reflex testing was performed on all consecutive non-serous OCs diagnosed at 1 academic hospital over a 2-year period; MMR deficiency was identified in 10/48 (21%) cases, all with endometrioid or clear cell histology. We conclude that there is a strong association between endometrioid and clear cell ovarian carcinomas and hereditary predisposition due to MMR gene mutation. These findings have implications for the role of tumor subtype in screening patients with OC for further genetic testing and support reflex MMR-IHC and/or MSI testing for newly diagnosed cases of endometrioid or clear cell ovarian carcinoma.
The pathogenesis of these microscopic extraovarian sex cord proliferations is unknown, but they may represent non-neoplastic proliferations of embryonic remnants.
Growing insights into the biological features and molecular underpinnings of ovarian cancer has prompted a shift toward histotype-specific treatments and clinical trials. As a result, the preoperative diagnosis of ovarian carcinomas based on small tissue sampling is rapidly gaining importance. The data on the accuracy of ovarian carcinoma histotype-specific diagnosis based on small tissue samples, however, remains very limited in the literature. Herein, we describe a prospective series of 30 ovarian tumors diagnosed using cytology, frozen section, core needle biopsy, and immunohistochemistry (p53, p16, WT1, HNF-1β, ARID1A, TFF3, vimentin, and PR). The accuracy of histotype diagnosis using each of these modalities was 52%, 81%, 85%, and 84% respectively, using the final pathology report as the reference standard. The accuracy of histotype diagnosis using the Calculator for Ovarian Subtype Prediction (COSP), which evaluates immunohistochemical stains independent of histopathologic features, was 85%. Diagnostic accuracy varied across histotype and was lowest for endometrioid carcinoma across all diagnostic modalities (54%). High-grade serous carcinomas were the most overdiagnosed on core needle biopsy (accounting for 45% of misdiagnoses) and clear cell carcinomas the most overdiagnosed on frozen section (accounting for 36% of misdiagnoses). On core needle biopsy, 2/30 (7%) cases had a higher grade lesion missed due to sampling limitations. In this study, we identify several challenges in the diagnosis of ovarian tumors based on limited tissue sampling. Recognition of these scenarios can help improve diagnostic accuracy as we move forward with histotype-specific therapeutic strategies.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.