Is Invasive Micropapillary Serous Carcinoma a Low-grade Carcinoma?

"Invasive micropapillary serous carcinoma" has been proposed as a synonym for low-grade serous carcinoma by some expert pathologists. In contrast, Singer and colleagues reported that some serous carcinomas with conspicuous invasive micropapillary pattern (SC-IMPs) can show high-grade nuclear atypia. However, the molecular features of such tumors have not been well documented. The aim of this study was to demonstrate and emphasize the fact that high-grade serous carcinoma confirmed by immunohistochemistry and molecular analysis can show conspicuous invasive micropapillary pattern. We selected 24 "SC-IMPs" and investigated: (1) their morphologic features; (2) the immunostaining pattern of p53 protein; and (3) KRAS/BRAF/TP53 gene mutations. The 24 SC-IMPs were subdivided into low-grade and high-grade tumors based primarily on the nuclear atypia, with the mitotic rate used as a secondary feature: low grade (n=5) and high grade (n=19). Low-grade SC-IMPs were characterized by low-mitotic activity, absence of abnormal mitosis, presence of serous borderline tumor, occasional BRAF mutation, and infrequent TP53 mutation. High-grade SC-IMPs were characterized by high-mitotic activity, presence of abnormal mitosis, conventional high-grade serous carcinoma, frequent TP53 mutation, and lack of KRAS/BRAF mutation. We demonstrated that high-grade serous carcinoma confirmed by aberrant p53 immunostaining and molecular analysis can show conspicuous invasive micropapillary pattern, validating Singer and colleague's report. Serous carcinoma with conspicuous invasive micropapillary pattern should not be readily regarded as low-grade serous carcinoma. Nuclear grade is the most important diagnostic feature in the SC-IMPs.

Ovarian high-grade serous carcinoma with a noninvasive growth pattern simulating a serous borderline tumor.

Ovarian serous borderline tumors (SBTs) being a precursor of low-grade serous carcinomas are morphologically characterized by noninvasive growth and low-grade cytology. On the other hand, many pathologists regard cytologically high-grade, noninvasive (HG-noninv) ovarian serous tumors resembling SBTs in low magnification as conventional high-grade serous carcinomas (HGSCs) by personal experiences. Nonetheless, there are no established molecular characteristic of such tumors. In this study, therefore, we attempted to provide the molecular evidence. We selected 37 ovarian serous tumors that exhibited a cytologically HG-noninv growth pattern, including 36 tumors that coexisted with conventional invasive HGSC components (HG-inv) and a single tumor exclusively composed of pure HG-noninv. Histologically, all HG-noninv showed many mitotic figures, and serous tubal intraepithelial carcinomas were identified in 3 tumors with HG-noninv. Immunohistochemically, most HG-noninv showed aberrant p53 expression, frequent IMP3 positivity, p16 overexpression, a high MIB-1 labeling index, and infrequent PAX2. By molecular analysis, the pure HG-noninv and 13 HGSCs with HG-noninv showed TP53 mutations, but KRAS/BRAF mutations were not detected in any of them. In 1 tumor, we detected an identical TP53 mutation in both HG-noninv and HG-inv components by using laser capture microdissection. These immunohistochemical and molecular features of HG-noninv were similar to those of conventional invasive HGSCs but different from those of SBTs. In conclusion, our results showed that a cytologically HG-noninv growth pattern simulating an SBT is a morphological spectrum of HGSC, but not a true SBT.

Ovarian transitional cell carcinoma represents a poorly differentiated form of high-grade serous or endometrioid adenocarcinoma.

Ovarian transitional cell tumors include Brenner tumors (BTs) and transitional cell carcinoma (TCC; non-BTs) according to the most recent World Health Organization classification. However, it remains a matter of debate whether TCC represents a distinct entity or a morphologic variant of high-grade serous adenocarcinoma (HG-SC). The purpose of this study was to resolve the above question by clarifying the morphologic, immunohistochemical, and molecular features of TCC. We reviewed 488 cases of epithelial ovarian carcinomas and reclassified them on the basis of the most recent World Health Organization classification with the modifications proposed by Köbel and colleagues, and 35 cases of TCC were identified; 25 and 6 TCCs were admixed with HG-SC and endometrioid adenocarcinoma (EC), respectively, and the remaining 4 cases were pure TCC. TCC components were not observed in any clear cell carcinomas or mucinous adenocarcinomas. Only 2 cases of malignant BT were identified. In addition to TCCs, malignant BTs, and related adenocarcinomas, benign and borderline BTs were included in the following immunohistochemical and molecular analyses. Immunohistochemically, pure TCCs, TCCs admixed with HG-SC, and pure HG-SCs were characterized by frequent aberrant p53 expression (diffuse or null pattern) and WT1+/ER+/PR+/IMP2+ immunophenotype, whereas BTs, including benign, borderline, and malignant BTs, were characterized by lack of aberrant p53 expression and WT1-/ER-/PR-/IMP2- immunophenotype. In contrast to the BTs, pure ECs and TCCs admixed with EC showed an ER+/PR+ immunophenotype. Nearly all the tumors with a TP53 gene mutation by molecular analysis showed aberrant p53 staining patterns. In conclusion, TCC is not a distinct entity but a poorly differentiated form of serous or EC, as (1) most TCCs coexist with HG-SC (mostly) or EC (occasionally), and (2) the immunophenotype and molecular features are similar to those of HG-SC or EC but different from those of BTs.