Morphological subtypes of ovarian carcinoma: a ... - BPA Pathology

SUBTYPES OF OVARIAN CARCINOMA 425changes. TP53 mutations have been demonstrated in some p53signatures and these may represent the earliest stage of developmentof high grade pelvic serous carcinoma. However, p53signatures are extremely common in the fallopian tube, even inpatients with benign disease and no hereditary predisposition todeveloping ovarian cancer, and it is clear that only a smallproportion will ever develop into a serous TIC. Serous TIC isnot diagnosed on the basis of p53 staining unless associatedmorphological alterations are present associated with a highMIB1 proliferation index.It is probable that not all high grade pelvic serous carcinomasare derived from the fallopian tube. A recent study whichsystematically examined all of the fallopian tubes in a consecutiveseries of ovarian carcinomas identified serous TIC innearly 60% of high grade serous carcinomas but not in othermorphological subtypes of ovarian carcinoma. 33 It is possiblethat some high grade serous carcinomas do arise from theovarian surface epithelium or the epithelium of corticalinclusion cysts, the latter being lined by ciliated epitheliumidentical to that lining the fallopian tube. Another possibility inthose cases in which no premalignant or malignant lesion isfound in the fallopian tube is that tubal epithelium mayexfoliate and become incorporated into the ovary and subsequentlygive rise to a high grade serous carcinoma. 33It can be summarised that there is accumulating evidence thatmany, or even most, high grade pelvic serous carcinomas (highgrade serous carcinomas which are currently classified asovarian, tubal or peritoneal in origin) arise from the fimbriaof the fallopian tube. In most cases, the malignant cellsexfoliate from the fimbria into the pelvis and abdomen andresult in the formation of an ovarian mass or masses usually, butnot always, with disease elsewhere in the pelvis and abdomen;this is conventionally referred to as ovarian high grade serouscarcinoma. In other cases, the neoplasm remains localised tothe fallopian tube, resulting in a fallopian tube high gradeserous carcinoma, or gives rise to extensive peritoneal diseasein the absence of significant ovarian or tubal involvement; thisis conventionally referred to as primary peritoneal high gradeserous carcinoma. It is likely that neoplasms which are currentlyclassified as high grade serous carcinomas of the ovary,fallopian tube and peritoneum are all different manifestationsof the same disease and the designation high grade pelvicserous carcinoma may be more appropriate. A consequenceof these observations is that screening programmes for ovariancarcinoma may be relatively ineffective in downstaging highgrade serous carcinomas since these are most likely disseminatedfrom the outset. However, screening may be of value inpicking up serous carcinomas when the burden of disease islower and will also identify other morphological subtypes ofcarcinoma which probably do arise within the ovary. Futurestudies investigating the underlying molecular events in thedevelopment of high grade pelvic serous carcinoma shouldconcentrate on the distal fallopian tube.MUCINOUS CARCINOMAPrimary ovarian mucinous carcinomas affect a wide age range,including occasionally children and adolescents. As discussed,they are relatively uncommon, the two studies referred toearlier suggesting that these account for approximately 3%of primary ovarian carcinomas, 7,8 a significantly lower percentagethan in older studies. The reasons behind the apparentmarked decline in primary ovarian mucinous carcinomas arewell known. In older studies, it is likely that many presumedprimary ovarian mucinous carcinomas, especially of advancedstage, were metastases from extraovarian sites. Advances inimaging, serum markers and preoperative workup have resultedin better recognition of metastatic ovarian neoplasms with theresult that many of these are not surgically removed. Moreover,pathologists are now better at recognising the morphologicalfeatures of metastatic mucinous carcinoma in the ovary, 1,46–52including the tendency to cystic change and the well knownmaturation phenomenon resulting in areas resembling benignand borderline mucinous cystadenoma. The use of differentialcytokeratin staining and other immunohistochemical markers53–60 has also improved the situation, although problemsstill exist. It is now clear that ovarian mucinous neoplasmsassociated with pseudomyxoma peritonei are almost always ofappendiceal origin, 61–63 with the very rare exception of primaryovarian mucinous neoplasms of overtly large intestinal typearising in a dermoid cyst. 64 It seems to be overtly large intestinaltype mucinous epithelium which has the potential to result inpseudomyxoma peritonei. There has also been a redefinition ofthe criteria for diagnosis of a well differentiated mucinouscarcinoma with so-called expansile (non-destructive, confluentglandular) invasion; 1,46–52,65,66 the distinction of this from amucinous borderline tumour at the upper end of the spectrum stillrepresents a somewhat poorly reproducible area amongst pathologists,resulting in some variation in the reported prevalenceof primary ovarian mucinous carcinoma between centres.Most primary ovarian mucinous carcinomas are unilateraland stage I and advanced stage neoplasms (stage III or IV) areextremely uncommon. In this scenario, a secondary shouldalways be strongly considered. Although metastatic mucinouscarcinomas in the ovary are still sometimes misdiagnosed as aprimary ovarian mucinous carcinoma or even a mucinousborderline tumour due to the prounced maturation effect seenwith some secondary mucinous carcinomas in the ovary, wehave to some extent come full circle in that, in my opinion,there is now a tendency to overplay the possibility of ametastasis even when the pathological features are obviouslythose of a primary ovarian neoplasm. I feel that in a largemajority of cases the distinction between a primary and secondarymucinous carcinoma in the ovary can be achieved bycareful pathological examination encompassing both the grossand microscopic findings and taking into account the distributionof the disease. It has been stated that when a mucinouscarcinoma is diagnosed in the ovary, further investigations,such as colonoscopy and detailed imaging of the upperabdomen, should be undertaken to exclude a primary neoplasmelsewhere but I feel this is unnecessary. Features suggesting ametastatic mucinous carcinoma in the ovary have been extensivelydiscussed elsewhere 46–52 and are listed in Table 3.Table 3Features favouring metastasis in ovarian mucinous carcinomaBilateral tumoursSmall tumoursNodular pattern of ovarian involvementMicroscopic surface deposits of tumourMarked lymphovascular space invasion (especially outside ovary and in hilum)Marked variation in growth pattern from one nodule to anotherDestructive stromal invasionSingle cell infiltration and signet ring cellsCells ‘floating’ in mucinExtraovarian spreadNone of these features are pathognomonic but are often seen in combination.Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.